Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2006-01-31

00:00:02elinenbeyou know what? this is the first time rockbox is working on a currently made player (the nano) since the archos recorder a few years back.
00:00:07BHSPitLappylinuxstb: how did they make it look all...graphical on that thread
00:00:13Bagderthe only thing with the currenyt system is that it has no "lower resolution" limit
00:00:21linuxstbBHSPitLappy: By using lots of bitmaps.
00:00:32elinenbeBHSPitLappy: you should download the H300 Optimized build from that site, and you can have the same screens...
00:00:46BHSPitLappyis using a bunch of bitmaps necessarily good practice?
00:01:04Bgernite all! amiconn, 10x for seeing my dumbness so quickly ;)
00:01:06linuxstbBHSPitLappy: I'm not an expert of WPSs - the iriver users at misticriver have built up a lot of expertise. Just don't tell them you're an ipod owner... :)
00:01:08BHSPitLappywell, I guess so, if retailOS does :P
00:01:08 Quit Bger ()
00:01:27linuxstbBHSPitLappy: How else can you create a graphical WPS?
00:01:46BHSPitLappylinuxstb: like podzilla2 does? gradients, shapes, etc?
00:01:57linuxstbBut that's slower.
00:02:11linuxstbBlitting a bitmap will be faster than drawing shapes.
00:02:23beeblehow fast is that grapgical wps?
00:02:38BHSPitLappyso what I understand now is that a .wps is just a layout for text and stuff
00:02:53BHSPitLappybut all that graphical-ness in that thread came from a separate feature?
00:03:33linuxstbIt's basically the same feature. A .wps file can refer to external .bmp files.
00:03:45BHSPitLappyah, I see that now
00:03:54linuxstbSome features like changing fg/bg colour, and setting margins on the display of text are not in CVS yet.
00:04:46 Quit Domonoky (Read error: 104 (Connection reset by peer))
00:04:50BHSPitLappydoes a wps take over the status bar?
00:04:53 Join topbloke [0] (
00:05:07BHSPitLappylike, replace it with its own rendition of that information?
00:05:11Paul_The_NerdBHSPitLappy: Only in that it can toggle the built in one not to display.
00:05:19linuxstbThat's optional - there is a tag to enable/disable the status bar.
00:05:28BHSPitLappyso then it can display its own battery meter, etc?
00:05:43BHSPitLappy(and be all pretty-pretty)
00:05:51linuxstbYes. Something like 10 different bitmaps depending on the current status.
00:06:50 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
00:07:07BHSPitLappyso...maybe my first one should be a retailos clone? :P
00:07:18linuxstbI think there already is one...
00:07:18BagderBHSPitLappy: I think there is one
00:07:36topblokehey sorry to interrupt but i got a Q: can i flash an old rockbox ajbrec.ajz file onto my archos?
00:07:37linuxstbA perverted iRiver owner created it...
00:07:44LinusNhmm, i can see a few possible optimizations in the wps code
00:07:44BHSPitLappyyeah, i think i see it
00:08:04Bagdertopbloke: you can flash rockbox at least
00:08:06BHSPitLappyit's really fugly though
00:08:10Furious_Gthanks a lot guys, im outta here
00:08:17LinusNthe most obvious is to not render the wps in the remote frame buffer if no remote is attached
00:08:29BHSPitLappyhaha, the winamp one is well-implemented
00:08:31topblokebut i only have the .ajz and not the source cause i modified it
00:08:39Bagdertopbloke: then you can't
00:08:43LinusNthe second one is to not render the wps in the main lcd frame buffer if the lcd is off
00:08:44topblokeaw :(
00:08:56amiconnLinusN: That idea goes in line with not rendering the wps on H3x0 when the backlight is off
00:09:11LinusNamiconn: that's the second one
00:09:18amiconnIt will be necessary to monitor status and do a full redraw on demand then
00:09:34LinusNyes, setting restore=true;
00:09:46topblokeso theres no way to make a flash file from an ajz?
00:09:59topblokehey man
00:10:01amiconntopbloke: No, why?
00:10:02LinusNif the peak meter is active, we can save quite some cpu
00:10:03topblokei'm back!
00:10:06midkaywb :)
00:10:14topblokefrom the dead!
00:10:22midkaydouble wb then!
00:10:23amiconnAll rockbox versions which support running from flash provide the needed .ucl file(s)
00:10:49topblokeok thanks
00:11:09peturso the peakmeter code can be disabled as well....
00:11:34topblokeoh and also is there a documented way to fix the headphone output - mine doesnt work on one side sometimes
00:12:06topblokei have a recorder
00:12:18 Quit Moos ("Glory to Rockbox !")
00:12:24 Quit petur ("here today, gone tomorrow")
00:13:17LinusNhmm, maybe a bool is_active() method in the screen_access api?
00:13:43amiconnEither that, or sending SYS_ events
00:13:59amiconnIt would still be necessary to read the initial status
00:14:34 Join Moos [0] (
00:15:37LinusNdoes lcd-h100-remote.c export the remote status?
00:17:39amiconnNo, but it has an internal variable tracking the status
00:17:57amiconnA readout function should be trivial
00:18:17LinusNwe could let the lcd drivers export the status
00:18:29amiconnYes, via a read function
00:18:42LinusNand map that to the display->active() function
00:19:03amiconnafaiu, the policy is to not export variables from firmware code
00:19:36LinusNthen the wps renderer could skip the rendering if display->is_active() == false
00:19:56 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
00:19:58LinusNthe only quirk is that it has to restore when the display becomes active
00:19:59amiconnYes, but it would need to track status itself
00:20:14topblokeok one more: If i flash a current rockbox ucl does it load a ajbrec.ajz from the root when it boots up?
00:20:22amiconn...and if last time display was off and is now on, do a full redraw
00:20:24Bagdertopbloke: no
00:20:28topblokesorry but i have been away from the scene for a while
00:20:34LinusNamiconn: yes
00:20:43topblokeok thanks
00:21:17 Quit Febs ("CGI:IRC (EOF)")
00:22:48 Quit chopped_pork ("CGI:IRC (EOF)")
00:22:55 Join webguest21 [0] (
00:25:34 Quit ender` (" Read what I mean, not what I write.")
00:27:23 Join RotAtoR [0] (
00:28:01 Join imphasing [0] (
00:30:03 Quit Moos ("Reboot")
00:30:55 Quit Kingstone (Read error: 104 (Connection reset by peer))
00:31:02 Join Kingstone [0] (n=sg1@
00:31:22 Join Moos [0] (
00:32:08linuxstbdamaki_, Furious_G : Are either of you still around?
00:36:03 Quit topbloke ("CGI:IRC (EOF)")
00:39:55 Quit DangerousDan (Read error: 104 (Connection reset by peer))
00:40:06BHSPitLappy"The images must be in a rockbox compatible format (1 bit per pixel BMP)" <- I'm guessing the 1bpp thing is outdated?
00:40:46damaki_linuxstb: yup
00:41:13kkurbjunamiconn: is there any way to compile a plugin without GCCOPTS, I'm trying to test and see if any flags are messing up what I'm working on, but every time I try and change the flags I get Incompatable model errors
00:41:35JdGordonhere.. there is nothing wrong with this line is there to open the damn file to write to.. fd = rb->PREFIX(open)(file,O_WRONLY|O_CREAT);?
00:42:23BagderJdGordon: you should simply use rb->open(file, O_WRONLY|O_CREAT);
00:42:29 Join _Kingstone [0] (n=sg1@
00:42:41 Quit Kingstone (Read error: 104 (Connection reset by peer))
00:42:46 Nick _Kingstone is now known as Kingstone (n=sg1@
00:43:07linuxstbdamaki_: Will you be able to test something in a few minutes?
00:43:53damaki_yes, for 15 minutes
00:44:34BHSPitLappyanyone know if the bmp coordinates in .wps begin at 0 or 1?
00:44:55 Quit DreamTactix291 (Read error: 110 (Connection timed out))
00:45:17damaki_0, I guess, as every C array should
00:48:02BHSPitLappyand transparent bmp's use magenta, right?
00:49:26linuxstbYes - #ff00ff
00:49:35BHSPitLappy255 0 255
00:49:40linuxstbThat as well...
00:49:53damaki_magic pink \o/
00:50:04BHSPitLappysome of us think in base 10 math, thank you very much
00:51:44CassandraIs transparency commited yet? I thought it wasn't working.
00:52:17LinusNworks just fine
00:53:04 Quit matsl (Remote closed the connection)
00:53:15BHSPitLappyoh, the bmp's need to be 16bit don't they
00:53:24BHSPitLappyis that why this 24 looks all messed up in the sim :P
00:53:31elinenbeLinusN: how is the single BMP for the WPS coming?
00:53:57elinenbeare you doing other rockbox related work as well?
00:54:23elinenbeanything exciting? :-)
00:54:44LinusNnot really
00:54:50Mooslike iaudio bootloader :-)
00:55:24LinusNMoos: soon
00:55:33 Join fairway [0] (
00:55:49MoosLinusN: I'm patient, and I know you're busy
00:56:13fairwayi have the problem that the ipod themes do not fit the full screen
00:56:15fairwayas seen here
00:56:25fairwayexcept for the status bar
00:56:32linuxstbdamaki_: I think I've found the problem with the lcd detection. But it may be 5-10 minutes until I can give you a build to test.
00:56:48Paul_The_Nerdfairway: That's because no themes have been made for the iPod yet. Those are all from devices with smaller screens.
00:56:51LinusNfairway: those themes were made for a smaller lcd
00:57:01fairwayhow can I adjust them?
00:57:05fairwayor did anyone else try one?
00:57:33fairwaydo you know of any themes made especially for the ipod?
00:57:33CassandraThe 'ipod' themes are designed to simulate an iPod on an iRiver H series.
00:57:44CassandraPossibly we need to change the names.
00:57:54linuxstbOr just delete them.....
00:58:12fairwayyes but did anyone made already themes for the bigger ipod screens?
00:58:32damaki_linuxstb: then I'll go to bed later, need... help open source... :P
00:58:32CassandraI would do, but I'm kind of busy with something else.
00:59:30fairwayi think it cant be too hard
01:00:27 Join Sinbios [0] (
01:00:41BHSPitLappythere aren't h300 themes already?
01:01:16Paul_The_NerdThere aren't any included with it yet.
01:01:49Paul_The_NerdMost of the h300 themes at misticriver require patches that haven't been incorporated yet (or have equivalent features incorporated)
01:01:51fairwayPaul_The_Nerd: yes but can we find them on the net?
01:02:26BHSPitLappyat misticriver?
01:02:42Paul_The_NerdIt's a website for iRiver fans. But again, *most* if not all of those require patches not in CVS
01:02:52Rob2222I think I have now a very _accurate_ rockbox runtime at H300s. 10,5h on a 1400mA stock battery.
01:03:05Rob2222Prooved by 2 new H300.
01:03:33 Join ashridah [0] (
01:03:36*JdGordon doesnt see such a big deal about batt life.. as long as it lasts maybe 4h im good
01:03:40 Join _cintia [0] (n=cintia_l@
01:03:49Rob2222More than one benchmarks made. And even benched without battery_bench plugin, same results
01:04:02***Saving seen data "./dancer.seen"
01:05:14 Quit fairway ()
01:05:32BHSPitLappymessing with making WPS files
01:06:06BHSPitLappyI used a 24bit bmp and the colors were all crazy, I used a 16bit one instead and it shows up as solid black
01:06:12BHSPitLappyany ideas?
01:06:39Paul_The_NerdIt should use 24 bit bitmaps.
01:06:55BHSPitLappyreally? to use on a 16bpp screen?
01:07:09Paul_The_NerdBut... maybe an endian issue again?
01:07:29LinusNthe bitmap loader doesn't yet support 8 and 16bit bmp's
01:07:48Paul_The_NerdLinusN: He tried 16 after the colors were horribly wrong on a 24, though.
01:07:52elinenbeLinusN: buy it now: $235
01:08:09XavierGrRob2222: What's the H300 runtime on the default firmware?
01:08:15BHSPitLappywhat's supposed to be a light grey to dark grey gradient, looks green and red and blue
01:09:17Rob2222XavierGr: I only made 1 bench with default FW, yet. Was with my cheap 2200mA battery. Runned 21h. RB runtime with that battery was 12,5h.
01:09:33Rob2222Same playlist, same volume, same MP3s
01:09:58XavierGrwow 9 hours difference?
01:10:13XavierGrMajor power leak, but where?
01:10:25LinusNXavierGr: the wps eats cpu
01:10:53Rob2222LinusN: It isnt deactivated when screen is black, yet?
01:10:58XavierGrLinusN: So disabling the wps (in a custom build) shoud justify 9 hours?
01:11:08Rob2222will do?
01:11:20LinusNXavierGr: probably not all, no
01:11:34XavierGrany idea for the rest of it?
01:11:59 Part _cintia
01:12:18XavierGrI mean 2 hours decline is acceptable but 9 hours is too much
01:12:28LinusNXavierGr: not really, that was what i was investigating when i fried my h320
01:13:16lostlogiclinuxstb: preglow: new version of the codec timer patch is posted on sf
01:14:03Rob2222LinusN: Do you have other power consumption compares? I made mine as accurate as possible.
01:14:34LinusNRob2222: no i don't
01:15:20 Quit Kohlrabi (Nick collision from services.)
01:15:26 Join Kohlrabi [0] (
01:18:39 Quit slarti (Read error: 104 (Connection reset by peer))
01:18:43 Join slarti [0] (n=tmartin@gentoo/developer/slarti)
01:19:21 Join kernel_sensei [0] (
01:24:03linuxstbdamaki_: Thanks for hanging around, but I don't think I'll finish my changes tonight. But I'm pretty sure I know how to solve the problem now.
01:25:42Rob2222but well, my only once made RB<=>IR compare isnt 100% perfect cause sthe battery started to rise its capacity a little.
01:26:04Rob2222RB benches I have enough so far, I will make more IR benches.
01:26:13damaki_right, then I'll surrender to my bed's call ;)
01:26:26damaki_good night
01:26:40 Quit webguest21 ("CGI:IRC (EOF)")
01:29:53 Join JdGordon41 [0] (
01:31:20linuxstbLinusN: I think there's a problem with the colour bmp reading function - it's not using the LCD_RGBPACK() macro, so won't work on the Nano and iPod Photo.
01:31:40 Quit JdGordon41 (Read error: 104 (Connection reset by peer))
01:31:47LinusNlinuxstb: it isn't?
01:32:18linuxstbNo, it seems to be doing the packing itself.
01:32:26LinusNah, yes
01:32:37LinusNcan you fix that?
01:32:39 Join JdGordon41 [0] (
01:32:42 Quit kernelsensei (Connection timed out)
01:32:45 Nick kernel_sensei is now known as kernelsensei (
01:32:59linuxstbLinusN: I haven't got anything set up to test it with.
01:33:08 Quit JdGordon41 (Read error: 104 (Connection reset by peer))
01:33:42linuxstbBut I could commit blindly.....
01:34:34LinusNyou don't even have an ipod?
01:34:58Rob2222I have some problems with the wiki. After preview and going back in my browser, my changes ar lost. :/
01:35:14 Join JdGordon41 [0] (
01:35:16linuxstbLinusN: I do, but no colour WPSs
01:36:11 Quit JdGordon (Read error: 104 (Connection reset by peer))
01:36:18linuxstbAnyway, I've fixed it, I can upload a patch if anyone can test it.
01:37:49LinusNlinuxstb: i just tried it in the sim and it works
01:37:50 Join DJ_TCK [0] (
01:38:32linuxstbYou mean you tried my patch 37 seconds after I uploaded it?
01:38:41stevebwhat are good linux tagging programs? anyone know?
01:38:59LinusNlinuxstb: no, i fixed it myself in exactly the same way
01:39:20LinusNworks on h300 target too
01:39:33linuxstbI say commit then. There's no reason it shouldn't work on the ipod.
01:42:19 Join webguest92 [0] (
01:44:59 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
01:46:14ashridahsteveb: i've heard reasonable things about a program called 'picard'
01:46:29*steveb has a look
01:46:49ashridahat least, i think that's the name
01:47:00stevebrings a bell
01:47:58steveboh its a music brainz thing
01:48:42steveblooks interesting
01:48:55CassandraWhat's the Nano screen size?
01:50:30CassandraNever mind. Found it.
01:50:55midkay176x132 !!!!!!!!!!
01:51:03CassandraPicard blows almost as badly as the other musicbrainz tagger.
01:51:04midkayi shall NOT never mind!
01:51:10CassandraIt' very much alpha code.
01:51:18CassandraPre-alpha even.
01:52:11BHSPitLappydon't listen to midkay, he's dead.
01:52:23midkaythen how could i be talking, fool?!
01:52:37midkaymy single cell grew to a multiquadrillion cell organism!
01:52:41CassandraAh, but aren't we all in a sense dead?
01:52:49midkayah, genius, genius.
01:53:46BHSPitLappyLinusN: w00t
01:56:48BHSPitLappyanyone know how to set the background color for a wps?
01:57:01Paul_The_NerdCan't yet.
01:58:10BHSPitLappyso...draw a great big white bmp then?
01:58:33Paul_The_NerdWell, if you did that, it'd draw over the text.
01:58:49BHSPitLappyyou can't draw any text over bmp's?
01:59:03LinusNBHSPitLappy: no
01:59:04Paul_The_NerdThat's what the transparent color is for.
01:59:20LinusNtime to sleep now, nite all
01:59:32 Quit LinusN ("Who is General Failure and why is he reading my disk?")
02:00:07Rob2222I was free to update the Iriver Runtime page:
02:01:50 Quit Moos ("Glory to Rockbox!")
02:03:37XavierGrRob: VBR?
02:03:43XavierGr128 VBR?
02:04:38Rob2222is it calles ABR?
02:04:57XavierGrvbr = variable bitrate
02:05:03XavierGrcbr = constant bitrate
02:05:11XavierGr128 is normally cbr
02:05:22XavierGrexcept if you changed it
02:05:52Paul_The_NerdXavierGr: ABR = average bitrate
02:06:18Rob2222No, I have reencoded my MP3s to 128kbit variable/average bitrate
02:06:34Rob2222Cause my 60GB disc was full.
02:06:48XavierGr128 vbr and full?
02:07:02Rob2222now I have 5GB free again. :)
02:07:15muesli__for the price of crappy q
02:07:36XavierGr128 cbr is a little crappy for my taste, imagine 128 vbr
02:07:54Rob2222muesli__: When I am in the city I didnt hear any difference.
02:08:10XavierGrI use mostly 192 alt preset standard or 160 cbr
02:08:30muesli__sure..but i wouldnt encode my music for any special occasion..
02:08:51Rob2222I have the original backups at my PC at home.
02:08:52muesli__XavierGr yepp
02:08:54paughRob2222: did you decode then recode at the lower rate or did you encode from source?
02:10:16Rob2222No I simply reencoded all MP3 with bitrates higher 130kbit. Yes I know that its quality overkill. But again, in my car or in the city with background noise I didn't hear the difference.
02:10:38Rob2222And I still have the original files.
02:10:41paughfair enough
02:10:57Paul_The_NerdThat's one thing nice about Vorbis. If the encoder's friendly, it does bitrate stripping, doesn't it?
02:11:50linuxstbAFAIK, it's been planned since the start, but never implemented.
02:11:57dpassen1bitrate peeling is a myth
02:12:21dpassen1Yes, it would be fantastic.
02:12:21Paul_The_NerdWhatever happened to it?
02:12:52dpassen1People have tried implementing it, and it always sounds much, much worse then a rerip.
02:13:09dpassen1Psychoacoustics doesn't scale down well, I suppose.
02:13:32linuxstbHow does lossy wavpack compare to other codecs? I rarely see it mentioned.
02:13:49dpassen1It's great at very high bitrates (300+ kbps)
02:13:51Paul_The_NerdI seem to remember someone having less than complimentary things to say about it.
02:14:34dpassen1It uses no psychoacoustics
02:16:42XavierGrRob: Do you keep your log files after the benchmark? (battery_bench.txt)
02:16:46dpassen1Once again, an idea better in theory then in practice.
02:22:20CassandraI still don't get why Rockbox takes way longer to install on the iPod than it does on the H140
02:22:58XavierGrinstall you mean copy the new folder and binary?
02:23:19linuxstbCassandra: As Paul_The_Nerd said, "disk mode" is much faster if you reboot into RetailOS.
02:23:39 Join DreamTactix291 [0] (
02:23:52CassandraIs it possble for Rockbox to do that by default?
02:24:15linuxstbYes, I wrote a patch to do that (Paul_The_Nerd runs it). But preglow didn't seem keen on committing it.
02:24:56CassandraAny idea why coot boting into USB mode doesn't work on the Nano? Hangs on the Rockbox USB screen
02:25:02Cassandracold even
02:25:05linuxstbIt's here
02:25:36linuxstbI haven't looked at that yet. It used to work, but suddenly stopped last week sometime.
02:28:26Rob2222I am confused. <= 30 minutes ago I measures the same 90V AC between my AC adapter plug and ground (power outlet). I checked that with 2 iriver AC adapters and a 3rd party 5V adapter. All had the same. Now I wanted to make another measuerment and the 90V AC agre gone. Only 5V AC now... ?!
02:28:30CassandraHmm. My colour bitmap doesn't seem to be loading. Do I need to save it in a special way?
02:28:51CassandraAnd can I load 2 layer bitmaps at the same time?
02:30:45XavierGrokay thanks. And be sure to keep them for a little while.
02:30:58XavierGrHave to sleep now. Later all.
02:31:14CassandraOh,another thing. I really dislikethe way the menu navigation keys are mapped to ff and rw. ay too easy to accidentally jog the jog wheel.
02:31:59CassandraUsing menu and select would work better, and be consistent with both the Apple firmware and the Archos player.
02:33:13Rob2222I dont like the ipod wheel at all. ^^
02:33:18CassandraWoo! Colour WPS!
02:33:26linuxstbCassandra: I could be convinced about that. I am used to the iriver way of using left/right to navigate the menus though.
02:33:28CassandraI am teh R3x0r!
02:33:31CassandraOr something.
02:34:02Cassandralinuxstb, I think the thing about accidentally jogging the jog wheel is the most convincing argument.
02:34:10BHSPitLappylinuxstb: I don't care too strongly, but yeah, that happens
02:34:16 Join YouCeyE [0] (
02:34:33linuxstbIt never happens to me. Maybe it's more of a Nano issue.
02:34:47BHSPitLappycould be.
02:34:57CassandraCould be. The Nano controls are smaller, of course.
02:35:09BHSPitLappydoesn't really happen to me too often, but it's kind of a design matter anyway
02:35:10linuxstbWhich would make it more sensitive to small movements.
02:35:30BHSPitLappythe wheel buttons weren't made to be used in a scrolling scenario
02:35:43linuxstbBut the problem we have is that "MENU" should be used to bring up the menu.
02:35:50linuxstbSo what do you do in the file browser?
02:36:14linuxstbUnless of course we don't map the MENU key to the menu...
02:36:17BHSPitLappy... what about it?
02:36:20Paul_The_NerdWell, what about the [Select] and >>| both being valid to choose menu entries?
02:36:23BHSPitLappyI don't see what you're saying
02:36:36Paul_The_NerdI mean, using |<< for back isn't really affected by jogging the wheel anyway.
02:36:55Paul_The_NerdThen, Menu stays how it is.
02:37:22CassandraI think we go for "long press on Menu to bring up menu".
02:41:15BHSPitLappythe dilemma is, in most cases, we want MENU to go back a level, but it's initial use causes it to go FORWARD in the sense of launching the menu...
02:41:47BHSPitLappyperhaps menu would just toggle between the menu root and filebrower, once you were at that level
02:42:01Rob2222iPod button mapping is a challenge... ^^
02:42:13CassandraPossibly we make long press on Select bring up the initial menu.
02:42:25CassandraWouldn't be the first time we've changed the function of a button.
02:42:34*BHSPitLappy just plain doesn't like long press
02:42:41CassandraI'm not so keen on overloading keymappings. We're kind of short on them.
02:42:49 Join linuxstb_ [0] (
02:42:59Rob2222long press for menu to bring up? mhh
02:42:59CassandraIt's less than ideal I know, but we aren't going to be able to avoid it.
02:43:18BHSPitLappywhat about my idea?
02:44:10Rob2222I only want to have the iPod Video display in my H360 then I would be very happy. :)
02:44:15BHSPitLappyfilebrowser >menu> menu root >select> submenu >menu> menu root again >menu> filebrowser again
02:44:25Rob2222And maybe iPod size. ^^
02:44:40linuxstb_BHSPitLappy: How do you move back a level in the file browser?
02:45:01BHSPitLappy....... prev? :D
02:45:38Rob2222select forward and menu back?
02:45:41linuxstb_Someone ask Apple to retrospectively add 4 more buttons to every ipod....
02:47:03BHSPitLappy"RECALL NOTICE: due to an unsatisfactory design for alternative iPod software, all units are being recalled for the installation of additional buttons. click here for more information."
02:47:20linuxstb_Sounds perfect.
02:53:14 Quit linuxstb (Read error: 110 (Connection timed out))
02:53:40 Nick linuxstb_ is now known as linuxstb (
02:54:23Paul_The_NerdAlright, with a very small change, the "Color settings 2" patch works on iPods, though it's kinda a roundabout way to go at things.
02:54:47midkayhahaha, BHSPitLappy
02:55:17Rob2222damn 3am
02:55:25Rob2222good night all
03:02:09 Join DJDD__ [0] (
03:04:05***Saving seen data "./dancer.seen"
03:11:01Paul_The_NerdCassandra: Did you say earlier you'd created a color Nano wps?
03:18:42BHSPitLappywps scripting is overly complex :(
03:23:19BHSPitLappyso... bitmapped wps's are really only good for one size of screen, right
03:24:02midkayhm, wondering if anyone here knows of any kind of app that would let you view statistics for a directory, more specifically i want to kind of index the files in there and see a list of most commonly used filenames?
03:25:27Paul_The_NerdBHSPitLappy: Basically, yes.
03:26:01BHSPitLappythen no compat worries :P
03:26:34 Join ehntoo [0] (
03:35:28 Quit San (Read error: 110 (Connection timed out))
03:43:38 Join ModernExecutive [0] (
03:44:01 Quit ModernExecutive (Client Quit)
03:46:04 Quit muesli__ (Read error: 110 (Connection timed out))
03:59:52Midgey34oh noes! an error
04:00:16Midgey3410 points for linuxstb
04:06:53 Quit mikearthur (Remote closed the connection)
04:13:09CassandraAnyone here know about the colour WPS patch?
04:14:05 Quit DreamTactix291 (Read error: 110 (Connection timed out))
04:14:13CassandraI have a colour WPS that's causing Rockbox to crash.
04:14:41CassandraI suspect it may be a full buffer problem, since it was working till I added a few more images.
04:14:53CassandraAnd I can't see anything different about those images.
04:16:18Paul_The_NerdDid you try removing an equal number of earlier images? Or better yet, an equal or greater amount in bytes of earlier images?
04:18:12CassandraIts easier just to bump up the buffer size and se if that fixes it.
04:18:26CassandraAt least it is if you have a compilation environment.
04:19:46Paul_The_NerdIs that a subtle "you don't, by chance, have one?" style question?
04:20:12Paul_The_NerdOr more of a "so that's what I'll be doing" style statement?
04:20:59CassandraI'm away from my development PC. ;)
04:21:19Paul_The_NerdDo you by chance know where the limit's set?
04:21:23CassandraAlthough I could do it remotely I suppose.
04:21:42Cassandrasomewhere in gwps_somethingorother.c
04:21:53Paul_The_NerdGimme a minute then
04:23:14CassandraI think so.
04:23:35CassandraIll find it in a mo when my source tree updates.
04:26:02CassandraHah, thought so.
04:26:18CassandraNo-one's multiplied the buffer by bit depth.
04:26:28Paul_The_NerdI was *just* about to say the same thing. Heh
04:26:57*Cassandra wanders what #define contains the appropriate magic number.
04:27:21Paul_The_NerdWhich appropriate magic number?
04:27:46CassandraLCD_DEPTH,I believe
04:28:12Paul_The_NerdI dunno
04:28:18CassandraAlthough if the image is stored internally as a 24 bit bmp, maybe not.
04:28:18Paul_The_NerdLCD_DEPTH on color targets should be 16
04:28:24Paul_The_NerdThe BMPs are 24, right?
04:28:51Paul_The_NerdSo, they make an assumption that the average bitmap will be no more than 1/10 the total area of the screen?
04:28:56CassandraNeed to have a look at the colour bmp loader.
04:30:46Cassandrapaul: No - the assumption is that the total number of bits is less than 1/10 of the 52 images.
04:31:07CassandraBear in mind that most of the bmps are only a few pixels across/down.
04:31:37Midgey34hmm, 1400 lines of code doesn't seem too bad for blackjack right?
04:31:38*Cassandra hopes the bitmaps are converted internally to LCD form before being stored.
04:31:49CassandraWaste of memory to save the whole lot.
04:34:32 Join Cassandra- [0] (
04:36:10Cassandra-Bad internet. No biscuit.
04:36:26Paul_The_NerdAll things considered, perhaps the biscuits are the problem?
04:43:48 Quit kernelsensei (Connection timed out)
04:44:54 Part YouCeyE ("Leaving")
04:46:20Cassandra-Agh. The bmp loader is hurting my head.
04:52:31 Quit Cassandra (Read error: 110 (Connection timed out))
04:52:31 Nick Cassandra- is now known as Cassandra (
04:52:52CassandraI'm just going to make the buffer 24bpp if LCD_DEPTH = 16 I think.
04:53:02CassandraIf it's inefficient, someone can fix it later.
04:57:36Paul_The_NerdI'm surprised that other WPSes didn't run into the limit already, actually. Like, with all the experimenters on MisticRiver.
04:58:04CassandraThe 1 in 10 pixel limit i obviously quite generous.
04:59:00Paul_The_NerdVery generous, but when you go from 1bpp to 16 or 24, it seems like it'd become pretty easy to get near the limit.
04:59:15CassandraYou know. I love technology. I love the fact that I'm sitting about 200 miles from my home workstation and using it to build Rockbox.
05:00:10 Quit LedFloyd ()
05:02:24 Join JdGordon [0] (
05:03:05 Join Cassandra- [0] (
05:03:22 Join Rob2222_ [0] (
05:04:09***Saving seen data "./dancer.seen"
05:08:48JdGordonCassandra: tech has far more annoyances than benefits :p
05:13:23Midgey34gah stupid senior quotes... I'm awful at this
05:15:49Midgey34in a less personal matter, I've posted a WIP blackjack plugin at MR
05:16:20Midgey34I'd especially appreciate iPod users opinions; the key definitions could use a lot of work
05:16:30Midgey34they seem awkward to me
05:17:17 Quit Paul_The_Nerd ("Leaving.")
05:17:36 Quit Rob2222 (Read error: 110 (Connection timed out))
05:19:09 Join Jungti1234 [0] (n=jungti12@
05:19:33 Quit Cassandra (Read error: 110 (Connection timed out))
05:19:33 Nick Cassandra- is now known as Cassandra (
05:20:06JdGordonhey Jungti1234
05:20:12Jungti1234hi JdGordon
05:21:20 Quit JdGordon41 (Read error: 110 (Connection timed out))
05:21:33 Join bradx3 [0] (n=brad@
05:51:50 Join DreamTactix291 [0] (
05:56:13 Quit bradx3 ("Leaving")
05:58:01 Quit Cassandra (" HydraIRC -> <- Leading Edge IRC")
06:06:55 Join linuxstb_ [0] (
06:08:40 Quit linuxstb (Read error: 110 (Connection timed out))
06:13:42 Quit DreamTactix291 (Read error: 110 (Connection timed out))
06:26:38 Join Membrillo [0] (
06:35:22 Quit RotAtoR ("zzzzz")
06:39:37 Join wubbla_ [0] (
06:52:23 Quit wubbla (Read error: 110 (Connection timed out))
06:57:57Jungti1234bad guys...
06:58:29lostlogicwow, just by rearranging some code in pcmbuf.c, I improved my new version's performance by several boost percent. Of course skipping is still totally broken
06:59:53lostlogicJungti1234: what's bad?
07:00:48Jungti1234lostlogic: They attack me
07:01:08Jungti1234Elementary students.
07:01:08lostlogicsadness... can't talk, must sleep. good luck :)
07:04:13***Saving seen data "./dancer.seen"
07:05:46 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
07:09:57 Join matsl [0] (
07:14:38 Join San [0] (
07:27:48 Quit ashridah (Read error: 110 (Connection timed out))
07:29:01 Join ashridah [0] (
07:31:20 Quit Jungti1234 ()
07:33:49 Join B4gder [0] (
07:35:24 Quit San (Read error: 110 (Connection timed out))
07:39:46Bg3rmorning :)
07:40:26Membrilloafternoon :)
07:41:03Bg3r;) someone should say now evening and the day will pass quickly :P
07:41:46Membrillowell, its a bit of a grey area here... 5:41pm. is that afternoon or evening? its still light
07:42:21ashridahMembrillo: it's just time to toss another snag on the barbie
07:45:38 Quit DJ_TCK (Read error: 104 (Connection reset by peer))
07:45:58Membrillowatch it boy :P
07:46:45ashridahwhat? barbeques are a part of our national tradition :)
07:47:05ashridahit's not like we call them weiners or anything :)
07:47:32 Join webguest94 [0] (
07:48:45 Quit matsl (Remote closed the connection)
07:49:23 Quit webguest94 (Client Quit)
07:52:33 Join DreamTactix291 [0] (
07:53:47Membrilloweeeeeiner dogs
07:55:32BHSPitLappyhmm this wps is acting strangely
07:59:41Membrilloashridah: i didnt know you were australian....
08:09:52MembrilloBHSPitLappy: whats it doing? can you make a screendump?
08:15:16Slasherilinuxstb_: i have now done the alignment fix patch for dircache. Would you like to try it? :)
08:15:19Slasherior preglow..
08:20:01 Join lamed [0] (
08:20:30lamedhi, guys. how do I tell make install to only build, say the plugins folder?
08:20:56 Join banan_ [0] (
08:21:17 Join thegeek_ [0] (
08:21:39 Join Jungti1234 [0] (n=jungti12@
08:21:41B4gderyou can't
08:23:08Jungti1234I want it.
08:23:16Jungti1234What should I edit?
08:23:20amiconnlinuxstb_: There's one RED in the build table...
08:23:54B4gder10 points for Dave
08:24:21amiconnGah, wiki spam :(
08:24:43B4gderoh that guy again...
08:25:19 Quit preglow (Remote closed the connection)
08:25:23 Join preglow [0] (
08:25:56Slasherithere really should be an image authentication when registering to wiki to avoid spambots..
08:26:37 Quit thegeek (Read error: 104 (Connection reset by peer))
08:26:47B4gderSlasheri: these aren't bots
08:26:56Slasherioh, that's weird
08:28:17Slasherithat some guys really have the time to abuse wikis by intention..
08:29:08B4gderwell of course I can't know this for sure
08:29:18B4gderbut I don't see any evidence of these being bots
08:31:36lamedamiconn "<amiconn> linuxstb_: There's one RED in the build table..." Thank YOU! now I got what "Red Builds" mean!!!
08:32:05B4gderdarn, the secret leaked
08:32:48lamedhehe :) b4gder: what _can_ I control over compilation with make?
08:33:16B4gderwhy don't you tell us what you want intead
08:33:37BHSPitLappyhi preglow
08:33:59 Join Paul_The_Nerd [0] (
08:34:28BHSPitLappy^ WIP, just started
08:34:39lamedI'm building brickmania for h100. I'm trying to put it in the sources file but A. I'm not sure what's going on in there and B. I'm waiting for a full setup in my archos directory for nothing. (p3 850mhz, that takes a while)
08:35:06JdGordonBHSPitLappy: needs colour :p
08:35:17BHSPitLappyJdGordon: talk to devs about that :P
08:35:54lamed(forget the "install" in my first question. how do I make just the plugins object files?)
08:36:32Paul_The_NerdBHSPitLappy: There's a patch that lets you set FG/BG color for the text.
08:36:54JdGordonlamed: run make then the compiled .rock should be in the apps/plugins/ folder
08:36:57BHSPitLappyPaul_The_Nerd: doesn't really do anygood, I'm making this more for others than myself
08:37:17 Join Alfafa [0] (n=hhj@
08:37:31lamedjdgordon: i'm trying to avoid make the whole thing, and learn something new
08:37:36Paul_The_NerdYeah, it seems unlikely to me that the final implementation will be very much like this patch.
08:38:14JdGordonlamed: i dont tihnk there is a way to just compile 1 .c at a time.. doing make only recompiles the stuff that need recompnig, so shouldnt take long after the first go
08:38:32 Quit merbanan (Connection timed out)
08:38:46AlfafaHow do I make a bug report for iriver h-120?
08:39:36Paul_The_NerdWell, there's a bugs section, and a forum. But you could start by just saying what's wrong, as maybe someone here knows a solution. :)
08:40:04lamedjd: what pc are you using?
08:40:07AlfafaI tried the rockbox firmware for the first time yesterday. It is quite good I think ;-)
08:40:23JdGordonlamed: POS celly 2.4, 512, winxp
08:40:40AlfafaThe problem is that there seems to be reintroduced a bug which also was in earlier versions of the original firmware
08:41:01Paul_The_NerdWhat bug, specifically?
08:41:06lamedjd: well, it shouldn't take a long time on _your_ pc... lol
08:41:17JdGordonhaha.. what u got?>
08:41:19AlfafaThe problem is when using the remote. There seems to be some background noise(not very loud) especially when it reads from the disk
08:41:54*lamed thinks Alfafa had a remote ticking problem using iriver fw !!!
08:42:00Paul_The_NerdAlfafa: That's actually a flaw in the hardware rather than software.
08:42:14AlfafaI bugreported it back then to iriver, and they somehow fixed. As far as I remember they changed the way the LCD on the remote was controlled or something
08:42:30lamedjd: p3 850 mhz. something like 250 mb of ram
08:42:36Paul_The_NerdAlfafa: If you go into the menu, choose General Settings, then Display, Remote-LCD Settings, there is a "Reduce Ticking" option at the bottom
08:42:42JdGordonhow long does a build take? 3min?
08:42:56lamedHEHEHEEE I wish
08:42:59AlfafaPaul_The_Nerd: Sorry. Haven't seen that. Thanks I will try that
08:43:06 Quit Jungti1234 ()
08:43:28AlfafaBut rockbox is nice :-)
08:44:35Paul_The_NerdIndeed it is.
08:44:56BHSPitLappynice like apple juice.
08:45:03Paul_The_NerdGah. Why do people give you *replacement* files, instead of patches?
08:45:06JdGordonand free like beer :D
08:45:52BHSPitLappyPaul_The_Nerd: because their modification is better than anyone else's?
08:46:17 Quit JdGordon (Read error: 104 (Connection reset by peer))
08:46:33lamedbecause they haven't read the how to work with patches manual?
08:46:40Paul_The_NerdVery possible, I suppose.
08:47:10Paul_The_NerdI was trying the blackjack game, and didn't realize it intended to overwrite my SOURCES file for the bitmaps. do i tell CVS to stop using the older version I wished it to retrive before..?
08:47:52B4gderupdate -A that is
08:48:10lamedyeah worked
08:48:34 Join JdGordon [0] (
08:49:21 Quit kernelsensei (Connection timed out)
08:51:47lamedi'm trying to figure out plugins/SOURCES file. how come sudoku.c is only under ipod Nano #if block, but it compiles for my h1x0 too?
08:53:57Membrilloi think its to do with ipod nano and h100 using the same defines or something
08:54:17*Membrillo has very limited C knowledge
08:54:25lamedno way, it's under | #if (LCD_WIDTH >= 220) && (LCD_HEIGHT >= 176) && (LCD_DEPTH == 16)| as well.
08:55:08B4gder#if (LCD_WIDTH != 176)
08:55:18B4gder /* These need adjusting for the Nano's screen */
08:55:25B4gdernow what can that mean?
08:56:24 Join webguest52 [0] (
08:56:43 Nick webguest52 is now known as GodEater (
08:56:50lamedwhoops. I was wrong. it's just under | #ifdef HAVE_LCD_BITMAP |
08:57:21lamedb4gder: was that a question or something to direct me in the right way?
08:57:37B4gderI find that snippet very clear
08:57:41Membrillorhetorical question :P
08:57:51B4gderif the lcd width is not 176, then sudoku is built
08:58:01B4gderand that is beacuse it hasn't been adjusted
08:58:14B4gderall that is more or less in plain english
08:59:18BHSPitLappybut it builds, doesn't it?
08:59:31B4gdernot if the lcd width is 176
08:59:47 Join Paul_The_Ner1 [0] (
08:59:55BHSPitLappywhoops, sokoban is what was on my mind
08:59:56lamedb4gder: yes, you're right. I just thought it's under | #if (LCD_WIDTH >= 220) && (LCD_HEIGHT >= 176) && (LCD_DEPTH == 16)| as well, i've missed the endif
08:59:59BHSPitLappydamn similar names.
09:00:04 Join einhirn [0] (
09:01:06B4gderI read the SOURCES file
09:01:23B4gderaround sudoku.c
09:03:13lameddon't you think it will be easier to read if all | #ifdef HAVE_LCD_BITMAP | plugins will be grouped under it, and after them all the other #ifdefs ?
09:03:57B4gderit might, yes
09:04:14***Saving seen data "./dancer.seen"
09:04:16lameddamn. my computer is so slow.
09:04:27B4gderand you use cygwin?
09:04:35 Quit Paul_The_Nerd (Nick collision from services.)
09:04:39 Nick Paul_The_Ner1 is now known as Paul_The_Nerd (
09:06:08 Join ender` [0] (i=ychat@
09:06:48Bg3ris there any possibility to make the flashes output sound under linux ?
09:07:10lamedyeah :((( I've installed ubuntu linux but after spending a few days _trying_ to set up an pptp connection unsuccessfully i gave up :(
09:07:36 Quit GodEater ("CGI:IRC")
09:07:51Bg3rflashes = shockwave flash
09:08:03B4gderI believe they sound at my place
09:08:07B4gderbut I'm not sure
09:08:13B4gderI *hate* web pages with sound
09:08:19ashridahBg3r: that just works for me.
09:08:46Bg3rashridah ok, but what does the swf plugin use for outputting sound ?
09:09:04Membrilloi hate anything that makes sound without me telling it to
09:09:59 Nick banan_ is now known as merbanan (
09:10:08ashridahBg3r: uh. same thing as most older apps. oss. possibly ESD if it's around
09:10:12ashridahalthough it should work without
09:10:27 Join ismo_ [0] (
09:10:59Bg3rah, so no way for arts ...
09:11:20ashridahperhaps with artsdsp
09:11:43Bg3rso u tell me to run firefox with artsdsp ? :)
09:13:35 Join LinusN [0] (
09:13:52ashridahi have no idea
09:13:59ashridahi use a sound card with hardware mixing
09:14:30B4gderCTCP sound is the invention of the devil
09:14:33B4gder== mirc
09:14:56BHSPitLappydid that actually output sound to people?
09:14:57 Join San [0] (
09:15:10Membrilloidk. i dont have my headphones on
09:15:12amiconnNot here
09:15:26Membrillo#rockbox sound
09:15:30Membrillonope doesnt work
09:16:06midkaythank you for ctcp sounding me.
09:16:17Membrillowhats ctcp?
09:16:24 Quit ismo (Read error: 113 (No route to host))
09:16:29B4gderClient to client protocol
09:16:37B4gderstuff in IRC
09:17:13amiconnI bet not many clients support ctcp sound
09:17:29Membrilloim on mIRC here
09:17:32midkaydidn't hear anything.. then again i didn't have speakers on...
09:17:37amiconn(HydraIRC obviously doesn't)
09:18:03B4gderbut it is so stupidly done
09:18:09B4gdersaying a file name to play
09:18:26Membrilloyou could easily replace it?
09:20:50BHSPitLappyamicon - using HydraIRC v0.3.148 :D
09:34:50 Quit San (Read error: 110 (Connection timed out))
09:43:28B4gderno ipod dailies?
09:44:12lamedwhere do I upload files quickly...?
09:45:11*B4gder detects the problem
09:45:53LinusNB4gder: i wish you could detect the problem with the daily installer binaries as well
09:46:14B4gderthose are evil
09:48:23lamedguys, can anyone adress me where could i upload some files, so i could put a comment in a sourceforge patch instead of making a new one?
09:48:43BHSPitLappyhuh, I just downloaded 1-30 for the nano
09:48:45 Quit DreamTactix291 (Read error: 110 (Connection timed out))
09:48:51BHSPitLappyif that's not a daily then I dunno what is
09:48:58B4gdertoday is 31st
09:49:13B4gderover here at least ;-)
09:49:31 Join Matze [0] (
09:49:55BHSPitLappyso the 1-30 build only goes through the 29th's changes?
09:50:19B4gderthe "1-30" build has all changes that was done when that package was made
09:50:25B4gderearly morning our time
09:50:57B4gderexcept for yday's build, since I did that manually
09:51:04B4gderat a somewhat later time
09:51:36 Quit lamed ("CGI:IRC")
09:54:01B4gderI've been pondering on separating the daily builds page from the cvs builds
09:58:51Paul_The_NerdWell, what is the argument against doing it?
10:00:50JdGordonhey, i had a crazy idea for rb, but i have no idea where to start so ill just put it here and hope some1 understands it...
10:00:54 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
10:01:59JdGordonwould it be possible to pause a plugin and return to the wps temporarily?? i noticed that the only button the h300 can do while the hold switch is on is rec and a.b, so that would be a good task switcher key...
10:03:08Paul_The_NerdI don't know if it's been tested, but for the logs, it seems like USB charging *is* working inside of Rockbox on iPod.
10:03:28MembrilloJdGordon: i dont know how you would do this but damn would it be useful
10:03:50JdGordoni thought of it while playing breakout and coming up to the end of the playlist
10:04:13JdGordonLinusN: would it be theoretically possible?
10:05:32LinusNtheoretically yes, practically a nightmare :-)
10:07:04Membrilloa lot of things are like that
10:08:15Membrillomaybe not a complete task switch, but how hard would it be to allow next\last track and vol+-?
10:08:54JdGordonna, access to the whole system, cept the plugins dir
10:09:06JdGordonplugin menu i mean
10:10:24 Join HaKuNa [0] (
10:13:37 Join safetydan [0] (
10:17:41LinusNJdGordon: we are discussing a way of creating a common plugin function for controlling playback
10:18:36JdGordonhmm... but more than that... ideally id like to be able to redo my playlist, and fiddle with settings and everything while breakout is paused in the background..
10:18:45MembrilloLinusN: terrific! this is the one feature i feel i would need of all of them
10:18:49JdGordonbut do understand the bitch it would be to accomplish
10:19:24Paul_The_NerdJdGordon: It'd probably be much easier to just have brickmania implement a save feature like several of the other plugins have.
10:19:45JdGordontrue.. but that was just an example
10:20:59 Join amiconn_ [0] (
10:21:50Paul_The_NerdBut, then, the other way Rockbox would be disallowing plugins the use of certain buttons.
10:21:54 Join DangerousDan [0] (
10:22:16Paul_The_NerdWhereas this way, the plugin creator decides whether they want their plugin suspendable (or an equivalent thereof) and how the user may go about doing so.
10:22:46JdGordonwell.. if hold is on then the rec/a.b button can trigger a button press but is dropped by rb...
10:23:04JdGordon(im talking h300.. dunno bout other targets)
10:23:19JdGordonjust a dream...
10:23:54Paul_The_NerdNothin' wrong with dreaming. It just seems like one that might be better implemented by plugin writers.
10:24:34LinusNthe idea would be to create a plugin library function, for example playback_control(), which would be like a mini-wps
10:24:57LinusNand the plugin author decides how to access it from his plugin
10:25:49LinusNor plainly reserving an odd key combination that all plugins must use, for example Play+Navi or something
10:26:02LinusNand the plugin default handler would handle it
10:26:38JdGordonnoo.. you couldnt reserver a key combo... there arnt enough keys to begin with..
10:26:44Paul_The_NerdOooh. I'm gonna be wandering around modifying all the existing plugins to let you control playback via the remote, then. :-P
10:29:07MembrilloI think a standard key combination would be wise. how many plugins do we have these days? if we leave it up to the plugin devs they could well just make it what they want and you would have to remember the different combination for each different plugin
10:29:30B4gderwe have to allow the plugins to do such things
10:29:42B4gderotherwise we limit the freedom to do things
10:29:47Paul_The_NerdYou could have a default behaviour, but I think that the plugin author should be able to decide to override it, if they need those keys.
10:29:56B4gderthat's a good idea
10:30:34Paul_The_NerdThat way, if the plugin author forgets to bother with choosing keys to control playback or whatever, in most cases it'll still be there anyway.
10:30:48Membrillobut then your going to get a billion people going "why can't i jump to WPS in blahblah plugin?!!?!?"
10:31:15B4gderpeople will always whine, there's nothing we can do about that ;-)
10:31:45Paul_The_NerdWell, are there any targets yet in which the hold switch is not dealt with in the software?
10:31:50Membrillowe can reduce it :P i mean what plugins use play+record
10:31:59 Join tucoz [0] (
10:32:08Paul_The_NerdI mean, the iPod already has very few buttons.
10:32:10B4gderwell, not all platforms _have_ a play+record
10:32:14Membrillogood point
10:32:28Membrilloim just looking at my h300 in front of me and typing ;)
10:32:37 Quit Maxime (Read error: 110 (Connection timed out))
10:32:59tucozHi, why not introduce a plugin menu? This could be accessed by a pre defined key-press and access to the above mentioned feature could be called from the menu.
10:33:14JdGordonthis is what we're talking bout
10:33:30tucozOh, sorry. I just read about different key presses.
10:33:39Paul_The_Nerdpreglow: ?
10:36:13preglowthis is going to be a lousy day
10:36:14preglowi can feel it
10:37:16Membrillohey. dont you talk about a lousy day. i had to go back to school after 9 weeks of holidays today :P
10:37:19 Quit amiconn (Read error: 110 (Connection timed out))
10:37:19 Nick amiconn_ is now known as amiconn (
10:37:49JdGordonpoor bugger
10:38:09JdGordoni beat you all tho.. ive been up since 5am (15hours and couting)
10:38:54Membrillo the other day i was sitting watching a dvd and i was like "wow, i havent eaten in 36 hours".
10:39:36 Quit linuxstb_ (Read error: 110 (Connection timed out))
10:39:37Membrillocant say ive ever said that to myself before. thats what holidays will do to you :)
10:39:47JdGordonthats terrible
10:39:49 Join linuxstb_ [0] (
10:40:14 Part tucoz ("Leaving")
10:41:17stamppotSlasheri: soo... tagcache? :-)
10:42:12Membrillodoes the h300 original firmware give preference to songs with higher playcounts in shuffle mode?
10:43:07*linuxstb_ makes an oops commit
10:43:12 Nick linuxstb_ is now known as linuxstb (
10:43:27LinusNlinuxstb: that's my job :-)
10:43:31*B4gder hands linuxstb 10 points
10:43:33 Part Alfafa ("Leaving")
10:44:00linuxstbLinusN: Well, you haven't done one for a while...
10:44:14B4gderlazy bastard!
10:44:18*B4gder runs
10:44:20*LinusN likes the blackjack plugin on mr
10:44:35B4gderyeps, that looks nice
10:44:40Paul_The_NerdWorks nice too. :-P
10:44:47LinusNand the code is goodlooking too
10:44:49 Join Jungti1234 [0] (n=jungti12@
10:44:51B4gdergood to see it being multi-platform from the start
10:45:00LinusNin fact, i don't have any complaints so far
10:45:06linuxstbDo we know the author?
10:45:17 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
10:45:18Slasheristamppot: when i have some time for it... :)
10:45:26LinusNTom Ross
10:45:54Slasherilinuxstb: hi, would you like the dircache?
10:45:54 Join DangerousDan [0] (
10:46:05Membrilloi hate people with first names as surnames :P
10:46:24linuxstbSlasheri: It would make all my dreams come true :)
10:46:25LinusNSlasheri: the dir cache event handling bug struck again
10:46:51LinusNit seemsit doesn't handle SYS_ events correctly
10:47:07linuxstbSlasheri: Do you have a patch for me?
10:48:10preglowcan anyone tell me what about the ipod's sound sucks?
10:48:16preglowi think it sounds completely decent
10:48:26BHSPitLappylinuxstb: (or any other ipod owners here) have you ever had headphone problems?
10:48:41BHSPitLappye.g. the famous right earbud crackle/failure
10:49:03BHSPitLappypreglow: sounds great to me
10:49:40preglowit's just as loud as the h120
10:49:43pregloweven for big headphones
10:49:50Paul_The_NerdIt's louder than the h120
10:49:54preglowso i bet the amp is rated the same
10:49:59MembrilloBHSPitLappy: ive only had problems with the white earbuds dying completely
10:50:09preglowmy white earbuds are broken
10:50:10Membrillothey just crackled out completely
10:50:19preglowcrackle at high volume
10:50:21preglowcompletely unusable
10:50:31Membrillopreglow: very common problem i think
10:50:36Paul_The_NerdI haven't used mine. I put them in my ears, and realized they weren't at all comfortable.
10:50:37preglowand, better yet, they're the only ones i have
10:50:38Membrillotheyre very very dodgy buds
10:50:39BHSPitLappyMembrillo: there's a large base of users whose headphones have been getting killed (right earbud) by the 5G
10:50:48BHSPitLappyand not just the stock headphones, expensive ones
10:50:55preglowhah, right earbud was the first to go for me as well
10:50:59BHSPitLappymy right earbud started to crackle today :(
10:51:01MembrilloBHSPitLappy: really? havent heard of that
10:51:07BHSPitLappyand I have a nano!
10:51:15preglowbut i only used them on my h120 :/
10:51:20Membrilloi dont use my ipod really. i lend it to a friend most of the time and use it for video on car trips
10:51:29 Join linuxstb_ [0] (
10:51:29Membrillomaybe ill use cheap headphones with the ipod now....
10:51:38Membrilloi have a g5
10:51:43preglowi don't see how an ipod can possible destroy headphones
10:51:48preglowi think this has been blown up somewhat
10:51:48stamppotmembrillo: so whats the use of buying one then in the first place? :)
10:52:40Membrilloi bought an ipod 4g, the screen shattered, i bought an h320, my friends house got robbed with the ipod inside, ipod got stolen, got an ipod video back on insurance
10:52:50Membrilloi dont use the ipod video
10:52:54linuxstb_The only unusual thing I can think of is that the ipod's headphone socket has a third ring with a video output.
10:53:03BHSPitLappynot the nano
10:53:03Membrilloso i lend it to the friend as a thank you for getting it stolen
10:53:06stamppotslasheri: no worries. I know what its like... If I had time to spare, I would have volunteered to help you out. But I've got a kid scuttling around and another one on the rise, so...
10:53:38stamppotmembrillo: :)
10:53:43linuxstb_BHSPitLappy: Are you sure? Why wouldn't apple wire it up on the Nano?
10:53:58BHSPitLappywhy would they?
10:53:59*Membrillo thinks all that the black jack plugin just needs to remember your last bet
10:54:08stamppotThe last line without context is a nice one: "So i lend it to a friend for getting it stolen..... :-)"
10:54:12linuxstb_BHSPitLappy: The same reason they have on the other ipods.
10:54:23Paul_The_Nerdlinuxstb_: Apple has said the Nano can't be used to display photos on TVs with the output set for the Color/Photo ipod.
10:54:32Membrillostamppot: hehehe yeah
10:54:32BHSPitLappylinuxstb_: why would the connector have the third ring on non-av-out-enabled ipods
10:54:40Slasherilinuxstb: yes, just a moment and i put it on web
10:54:47Paul_The_NerdThough that doesn't mean it won't, one day.
10:54:55SlasheriLinusN: yep, i saw that also.. i will figure a way to fix that soon
10:55:00linuxstb_BHSPitLappy: OK, sorry. I just assumed the Nano had video-out capabilities as well. Seems it doesn't.
10:55:13BHSPitLappynano's pretty much empty :P
10:55:21BHSPitLappyno recording, no av-out
10:55:56Slasherilinuxstb: hmm, or i try dcc first if that works
10:56:06BHSPitLappyI wonder if linux will figure out a software-based analog video-out driver someday :P XD
10:56:22Paul_The_NerdBHSPitLappy: It's not *definite* that it won't be able to record, I thought.
10:56:33BHSPitLappyit just plain doesn't have the hardware
10:56:40Slasherilinuxstb_: good, seemed to work fine. i will then commit it if that works :)
10:56:49linuxstb_Slasheri: Yes, testing now. Thanks.
10:57:30linuxstb_Slasheri: Maybe that define should be something like ROCKBOX_STRICT_ALIGN
10:57:30Paul_The_NerdBHSPitLappy: I thought it could potentially record via the serial using the connector?
10:57:46*Membrillo just found that you can't actually lose on blackjack. Membrillo just beat the computers 17 with a 12
10:58:01*BHSPitLappy shrugs
10:58:05amiconnA general method to intercept plugins will be a real nightmare imho
10:58:05Paul_The_NerdMembrillo: I can lose plenty....
10:58:18amiconnSome plugins do stuff that can't be easily intercepted
10:58:23BHSPitLappyPaul_The_Nerd: if they prove me wrong, come back and laugh at me... I won't care, I'll be too busy recording crap then :)
10:58:23amiconn...e.g. grayscale
10:59:13Paul_The_NerdBHSPitLappy: I have an iHP-120 anyway. It's not of any concern to me, in the end.
11:00:10Slasherilinuxstb_: hmm, true
11:00:13Slasherithat sounds better
11:00:35*BHSPitLappy Human - Metallica - 128kbps - iPod Nano - Rockbox!
11:00:53BHSPitLappy(AAC, no studder)
11:01:06LinusNamiconn: then those plugins would not call the default handler, but handle the key combo explicitly
11:01:17LinusNamiconn: i can't see the "nightmare" in that
11:02:18ashridahthat sounds about right to me. make it *easy* for plugins to do the default action, but if they don't explicitly do it, then you don't second guess them
11:02:42Membrillowhat do people advise i do in this sitch. I'm halfway through listening to an album, i want to change albums after the song im currently listening to has finished. so i want to play a different folder once this song has ended and have that new album as the only album in the playlist. is there a way to do this other than wait till the end of the song and create a new playlist?
11:03:10LinusNif a plugin that doesn't handle the key combo, the default handler will handle it
11:03:28ashridahMembrillo: you can add stuff to the playlist as 'next'
11:03:44Bg3rMembrillo hold on the navi button
11:04:03Membrilloyeah, but then my playlist has the new album in the middle of the old album. i want to start listening to the new album on repeat.
11:04:09LinusNso now we need a plugin_wps() plugin library function
11:04:17linuxstb_Slasheri: Wow....
11:04:19***Saving seen data "./dancer.seen"
11:04:30Bg3rMembrillo go the the new album's folder, hold navi on it ... playlist->insert last
11:04:50ashridahBg3r: he wants to wipe out the remaining entries in the playlist first
11:04:58 Quit linuxstb (No route to host)
11:05:09ashridahso it ONLY plays the new album, and then stops
11:05:29Membrillowell, only plays the new album on repeat
11:05:31linuxstb_Slasheri: dircache seems to be working perfectly. And it took less than a second to build it - the debug screen says "Scanning tool: 0s".
11:05:38Membrilloi want to change albums and listen to the new one on repeat
11:05:42Membrilloim sick of the old one
11:05:45Slasherilinuxstb_: huh, that's nice :D
11:05:48Membrillobut i want to listen to the rest of the current song
11:06:00Slasherithen it must have a really fast disk..
11:06:33Slasherilinuxstb_: i will just commit that then
11:06:34linuxstb_I don't have _that_ many files on it. Let me just count them.
11:06:44Membrillono solution other than wait till the end of the song and select the new one?
11:06:55dwihnoIs there a C call to execute a program and then return?
11:07:12ashridahMembrillo: i just don't know how to delete selected entries from the playlist.
11:07:12LinusNdwihno: that depends on the OS
11:07:15amiconnLinusN: So how would you return from wps to plugin?
11:07:17linuxstb_Slasheri: Commit away.
11:07:24ashridahand if there is such a feature, idon't know if there'll be a quick way to do that
11:07:29LinusNdwihno: look for system()
11:07:53LinusNamiconn: i'm not talking about the internal WPS
11:07:59Membrillohmmm, i just want to clear the playlist (including the currently playing song) without the song stopping. can you play a song without it being in a playlist?
11:08:01linuxstb_Slasheri: I've got about 1900 files/directories on my disk.
11:08:05LinusNi'm talking about a mini-wps in the plugin libarry
11:08:36LinusNi'm not letting the plugin go to the normal wps
11:09:03dwihnoLinusN: I read something about exec and friends, but the do not return to the software, no?
11:09:13LinusNdwihno: true
11:09:23Slasherilinuxstb_: hehe, anyway that's fast :)
11:09:30ashridahdwihno: sort of.
11:09:39ashridahthey replace the execution image
11:09:49linuxstb_Slasheri: Yes, but the spinup time was slow - it's great to have dircache back :)
11:09:56ashridahyou typically use them in conjunction with 'fork()' if you want them to 'return' to the parent
11:10:22dwihnoLinusN, ashridah: ah, I'll look for the system() call then
11:10:28ashridahbut then you're dealing with two separate processes. you can resync them by just issuing 'wait()'
11:10:52ashridahbut yeah, system() is the shortcut method
11:11:10ashridahit's basically a combination of fork, exec and wait
11:11:15amiconnSlasheri, linuxstb_: Imho it would be better to always use correct alignment
11:11:31amiconnWhile not strictly necessary for coldfire, it helps performance
11:11:47LinusNand less clutter in the code
11:11:59preglowhow large an impact does it have on data size?
11:12:05Slasheriamiconn: with dircache that would only increase space it takes..
11:12:20Slasherii doubt there would be any performance boost
11:12:29preglowSlasheri: btw, you can replace most of those CONFIG_CPU checks with defined(CPU_ARM)
11:12:38Slasheripreglow: ah
11:12:45preglowat least the PP ones and the PNX one
11:13:46preglowSlasheri: but yeah, how much space is wasted when alignment is forced?
11:14:31dwihnoashridah, LinusN: thanks, I got it working now. the exec stuff wasn't just right for this project.
11:14:43ashridahexec's complicated
11:14:48 Quit safetydan ("Leaving")
11:14:49ashridahdealing with arguments and whatnot can be a pain
11:15:21ashridahbut it gives you the option of doing stuff with the processes' stdin/stdout and stderr, and allows things to run asynchronously
11:15:59 Join San [0] (
11:16:12Slasheripreglow: it depends, 0-3 bytes per entry.. so about 5-20 kB in total.. not so much in fact :)
11:17:51preglowthis isn't my idea of cluttered code
11:18:06preglowso you could keep unaligned mode for targets which can do it for my sake
11:18:39Slasheriyep, there is only one #ifdef for that in the code
11:18:46preglowyes, exactly
11:19:05preglowwell, i've gotta go move some stuff
11:23:59 Join DreamTactix291 [0] (
11:29:59 Join Siku [0] (
11:33:18prhjust got to say - rockbox is awesome!
11:38:37 Nick Rob2222_ is now known as Rob2222 (
11:38:47Rob2222good morning
11:39:49Rob2222LinusN: I got an answer. He will look after a PCB, but he coudln say something, cause he sayed that H300 spare parts always were rare.
11:40:18LinusNRob2222: thanks, let's hope for the best
11:40:57Rob2222Yes. But I thought, when you buy a player and then sell the HD, it should save money, too.
11:41:36LinusNspare parts are always nice to have
11:41:46Rob2222hehe ok :)
11:43:04linuxstb_Has anyone made any changes recently to the Rockbox USB detection to do something special if USB is plugged in whilst Rockbox boots?
11:43:41LinusNnot that i know of
11:44:15LinusNexcept for the usb charge mode in h300
11:44:20linuxstb_Mmm. So I can't blame anyone but myself for it being broken on the ipod...
11:44:42linuxstb_LinusN: Yes, I can't see anything with your commits that could affect any other targets.
11:45:20BHSPitLappywell, it executes the code to show the usb logo
11:45:37BHSPitLappyshouldn't that be lumped together with the reboot function
11:46:20linuxstb_BHSPitLappy: It is, but it only works when you plug in USB normally. If you plug in USB with the ipod turned off, then Rockbox will start, and go straight into that USB screen.
11:46:30linuxstb_(instead of rebooting back into disk mode)
11:47:15amiconnThis is early USB mode, invoked before initialising ATA
11:47:16BHSPitLappyi know
11:47:27BHSPitLappyjust stepping back and analyzing this :)
11:47:30amiconnMaybe you need to hook into that separately
11:47:30linuxstb_amiconn: Is that new?
11:47:42amiconnIt's a year old or so
11:47:50linuxstb_Mmm. Strange, because it seemed to work in the past. It only broke recently.
11:48:30linuxstb_amiconn: Does that still call the usb_enable() function?
11:51:06damaki_good morning
11:51:13 Nick damaki_ is now known as damaki (
11:52:17linuxstb_amiconn: I've found it now - looking at the code, I've no idea how it could have worked in the past...
11:53:07amiconnIt does call usb_enable()
11:54:20 Join NightCat [0] (
11:54:26linuxstb_Does it? From where?
11:54:36NightCatHi all!
11:55:16amiconnlinuxstb_: The USB thread does when all threads acknowledged USB
11:55:40amiconnThe final acknowledge (for the main thread) is done in usb_screen()
11:56:01amiconn...but now I remember having problems with early USB on iriver as well
11:56:32amiconnPerhaps some thread doesn't properly acknowledge USB at this point?
11:56:34NightCatI have a question: does somebody knows where can I get some more fonts for RB? (I want some more fonts with Russian support...)
11:56:35linuxstb_Didn't Slasheri add some more threads recently for audio initialisation and such like/
11:56:50amiconnUsually no problem, because we have bootloader USB on iriver
11:59:09 Join aegray_ [0] (
12:01:19Slasherilinuxstb_: not more threads, i just changed the audio initialization order
12:01:39Slasheriso now the audio thread is initialized earlier so it can initialize pcm hardware on background
12:01:41NightCatHm, what about fonts? ;)
12:01:59linuxstb_Slasheri: Does it respond to usb events?
12:02:21Slasheriusb is connected very early
12:02:31Slasherithen it might hang
12:02:33linuxstb_Which it is....
12:02:50linuxstb_If USB is plugged in before the ipod boots, it hangs on startup.
12:02:54Slasherihmm, yep.. i will think how to fix that
12:03:01Slasheriah :/
12:03:31Slasheriit hangs waiting for the correct event (Q_AUDIO_POST_INIT).. it don't accept any other event before that
12:04:58linuxstb_At least on the iPod, it should accept a USB event - this will cause the iPod to reboot anyway, so it doesn't matter that the audio is half-initialised.
12:05:34amiconnSlasheri: The dircache thread also has a bug
12:05:50amiconnIt stops scanning if it receives _any_ event
12:06:09amiconn..even if it's such a minor thing like charger plugged/unplugged
12:08:06Slasheriyes, i know
12:08:42Slasherii will try to figure some good way to fix these issues
12:12:50 Quit aegray (Success)
12:13:43linuxstb_damaki: Have you tried the latest bleeding edge Rockbox build (and the latest bootloader-color.bin I've uploaded to the installation page) ?
12:14:11 Join webguest67 [0] (
12:14:17damakinot yet
12:14:40damakii'm gonna try it now
12:14:54linuxstb_Thanks - it should hopefully solve your lcd problems.
12:15:07NightCatAsking again :) Is there some way to translate simple system fonts (from windows) to RB format?
12:15:23B4gderNightCat: if you can convert them to bdf, yes
12:15:57NightCatI want some more fonts...
12:16:28 Quit uarx ("Leaving")">
12:16:53NightCatHaha, I find converter :)
12:17:31NightCatB4gder: cirilic means russian?
12:17:46NightCatOh, yes, I see :)
12:17:52NightCatThanks :)
12:19:43Membrillo what do people think about this. Its the source for a FNT2BDF converter (in C)
12:20:02B4gderis fnt a bitmap format?
12:20:16NightCatAnd what about this:
12:20:32Membrillowait, are rockbox fonts in BDF format?
12:20:44Membrillowhat are rb fonts in?
12:20:47B4gderbut convbdf converts from bdf to the rockbox format
12:21:06B4gderrockbox fonts are using rockbox format ;-)
12:21:15NightCatWhere to download "convbdf"?
12:21:17linuxstb_Is bdf a bitmap format?
12:21:25B4gderlinuxstb_: yes
12:22:14Bg3rbitmap description font ?
12:22:21Membrilloso could you integrare the thing i linked into convbdf and have fnt->bdf->rb?
12:22:22preglowlinuxstb_: oh, i forgot to mention
12:22:34preglowlinuxstb_: usb mode worked for me again yesterday when i did my adjustable cpu freq tests
12:23:01preglowlinuxstb_: so something weird is definitely afoot
12:23:07XavierGrHehe I've got in front of me the box of a new H340! Now lets open it to see what's inside!
12:23:10linuxstb_preglow: Are you talking about "normal" usb mode - i.e. plugging in usb when Rockbox is already running?
12:23:24preglowthe reboot part
12:23:29preglowthat doesn't work for me anymore
12:24:37linuxstb_Has anyone else mentioned problems with the normal usb mode. I thought that generally, it was only the early usb mode that was broken.
12:24:51damakilinuxstb_: I have a question : is there somewhere a screenshot with the good colors. I just realized I do not know what the colors should be. I say that because the rockbox splashcreen is good but I must admit that I do not know about the other colors. It could be a screenshot or just the default wps hex colors, so that I could compare
12:25:17linuxstb_The default wps is black on a light blue background.
12:25:43linuxstb_The Rockbox logo can be seen at - if the logo is fine, everything else should be.
12:25:48damakiand what about the text highlight ?
12:26:03linuxstb_That's a sort of dark brown.
12:26:09Rob2222XavierGr: Congratulation.
12:26:36linuxstb_damaki: If you go to Menu -> Info -> Version you can see the Rockbox logo again.
12:26:41Rob2222XavierGr: Now you have 5 minutes to install RB. ;)
12:27:12XavierGrStrange though advqancedmp3players advertised it as the SE version. It has both an lcdless remote and a cradle!
12:27:40damakithen the colors are good and I was partially wrong yesterday. the highlight color was red
12:27:50XavierGrThe funny part is that there is no stock left for H300s on advancedmp3players anymire
12:28:01NightCatXavierGr: I have an lcdless remote, but not craddle...
12:28:49XavierGrI have the cradle in front of me :)
12:28:51NightCatXavierGr: 4 minutes left for installing RB :)
12:29:09XavierGrHow do I turn it on. Do I have to charge it first?
12:29:22NightCatI think - no :)
12:29:34NightCatJust push THE BUTTON :)
12:30:01XavierGrI remember that the h140 I had was middle charged...
12:30:04linuxstb_damaki: Have you upgraded to the latest official bootloader and now then?
12:30:19Rob2222All Li-Ions should be middle charged.
12:30:19JdGordonany1 know what font the family guy wps needs to look proper?
12:30:27 Quit Membrillo ()
12:30:45NightCatJdGordon: What?
12:30:54damakiyup, and the lastest bleeding age
12:31:23linuxstb_damaki: Good, that's working then. Thanks.
12:31:29NightCatWhere the USB mode picture is in source?
12:31:30XavierGrstrnage it will not power up! I will try to charge it
12:31:36damakilinuxstb_: Thanks too ;)
12:31:37linuxstb_If the logo is fine, then everything else will be.
12:32:00XavierGrArgh the AC adpater has a UK plug! F*ck!
12:32:07NightCatXavierGr: It's just broken haha :) (Just kidding.)
12:32:10XavierGrNow where is my adapter
12:32:43Rob2222XavierGr: The H140 adapter does NOT fit.
12:33:31NightCatI'm asking AGAING!!! DOES Anybody knows where the usb mode logo is in source?
12:33:35 Join mikearthur [0] (
12:33:51B4gderyes we know
12:34:18NightCatHaha :) Where?
12:35:00JdGordondo plugins have proper access to the hdd? i.e move/rename files etc?
12:36:18Rob2222Hmm, a Norton Commander Plugin would be nice. :)
12:36:20BHSPitLappybejeweled saves highscores
12:36:39B4gderJdGordon: yes they do
12:37:01Rob2222JdGordon: Thinking about to move files?
12:37:13JdGordonim tihnking like u am
12:37:36LinusNwee, cache-aligned asm-optimized ata read loop @124mhz: 7.8mbytes/s
12:37:47JdGordonum... woot
12:38:09JdGordonsoo.. uh.. thats good right? :D
12:38:20linuxstb_LinusN: How do you test the read speed?
12:38:27Jungti1234color = colour?
12:38:31LinusNa benchmark plugin i wrote
12:38:46B4gderJungti1234: english vs american
12:38:57BHSPitLappybackwards :P
12:39:05LinusNamerican vs english
12:39:09BHSPitLappyJungti1234: england likes "u"s :)
12:39:19linuxstb_LinusN: Do you think that's suitable for CVS? Could be interesting for people to compare different hardware.
12:39:23LinusNand england doesn't like US :-P
12:39:38B4gderuuuuh, that hurt
12:39:47LinusNlinuxstb_: i'll consider it
12:40:13BHSPitLappyLinusN: sure didn't act like that 200 years ago
12:40:31XavierGrHaha I just listened again the "iRiver Catch the digital flow"! I am moved, I remember the same thing with my newly bought iHP-140!
12:40:52Rob2222XavierGr: How do you got power? :)
12:41:27XavierGrI used a UK to EU converter
12:41:34Rob2222Ah OK. :)
12:41:40XavierGrAlso does it have to be lit when charging?
12:41:44Rob2222XavierGr: Where are you from?
12:42:02Bg3rXavierGr happy rbing on h300 :P
12:42:24Rob2222XavierGr: Yes it can, no problem. You can make ALL while charging.
12:42:45Rob2222Should I send you 1.29 kr patched wia DCC? :)
12:42:45XavierGrThanks, but I will fiddle a little while with the original firmware to see differences between H100.
12:43:03XavierGrWhy korean?
12:43:13XavierGrIsn't 1.30 the latest?
12:43:27Rob2222You can set the FM frequency in the KR firmware.
12:43:41XavierGroh interesting
12:43:48Rob2222=> you could listen to police here
12:44:01NightCatWhy this fonts:"> doesn't show russian symbols?
12:44:26XavierGrAlso now I will have to make a modded cable for the cradle.
12:44:34Rob2222XavierGr: 1.30 has removed video support I think.
12:44:43XavierGrI don't want to use the AC adapter for that.
12:44:44Rob2222Modded cable?
12:44:54Rob2222What you want to use?
12:45:12Rob2222But not a USB plug, or?
12:45:21 Quit HaKuNa ("CGI:IRC")
12:45:29XavierGrI will leave the cradle on my desktop and use 2 usb ports. One for power and one for synching
12:45:32Bg3rRob2222 ????????
12:45:36Bg3rremoved video support ?
12:45:50Rob2222Bg3r: In 1.30 I thought.
12:45:55 Quit paugh ("Leaving")
12:46:21Rob2222XavierGr: 1 USB connector is to less to support the power plug!
12:47:11Rob2222Bg3r: Am I wrong? As I said, "I think".
12:47:21 Quit Siku ()
12:47:27Rob2222XavierGr: USB = max 500mA
12:47:36XavierGrH140 seemed to work just fine both charging and synching, not to mention only charging
12:47:41Rob2222XavierGr: Iriver charging and running consumes a lot more
12:47:43 Quit ehntoo ("Leaving")
12:48:02XavierGrRob: I doubt that it needs more than 500mAh when charging only.
12:48:14Rob2222Depends on your motherboard layout ;)
12:48:33XavierGrBut you did the measurements, how much it needs to charge?
12:48:47B4gderit generally needs more than 500mah to charge fast, yes
12:48:56Rob2222The H300 is able to charge the battery with up to 650mA
12:49:19Rob2222Accoring the LX2201 battery charger datacsheet.
12:49:20XavierGrAh yes, thruth is that my H160 needed more that normal time to charge in USB.
12:49:26JdGordonSan: i love your family guy wps
12:49:31XavierGrBut no worries for that, as long I will not damage it.
12:49:51XavierGrSo you think I might damage it?
12:49:53Rob2222XavierGr: You can only damage yoz motherboard USB connector.
12:49:53NightCatMy AC adapter says: 5V 2A...
12:50:02Rob2222XavierGr: Depends.
12:50:13XavierGrmine too, both H140 and H340
12:50:42XavierGrBut as I said: I have made a dock for my H140 that takes power only from 1 USB port.
12:50:42NightCat2A - cool, it can kill a man :)
12:50:51XavierGrIt manages to synch and charge though it will not charge it fast.
12:51:08Rob2222XavierGr: USB says, that every USB connector must have at least 500mA. I read that some motherboard companies support much more current at their usb plugs. But I cant say how you can test this.
12:51:12ashridahNightCat: uh, dude, a few hundred milliamps can kill a man
12:51:26ashridahonly half a dozen milliamps can kill someone if you pierce the skin
12:51:26XavierGrNow I will make a cable only for power and one for synching in order not to stress my motherboard
12:51:27NightCatYes :)
12:51:35 Nick linuxstb_ is now known as linuxstb (
12:51:43XavierGr(though my motherboard seems to be a bear in those things)
12:52:32XavierGrSh*t 2 more cables in my desk. In a while I will swim between cables....
12:52:55Rob2222XavierGr: Simply get 5V over at HD connector from the PC power supply.
12:53:08B4gder"it doesn't work at all" is a great bug report... :-)
12:53:32NightCatWho wrote this?
12:53:38XavierGrRob you mean inside the box from an IDE connector?
12:53:43B4gdersee bug tracker
12:54:05XavierGrIDE = not the ribon cable but the power connector
12:54:20linuxstbB4gder: No, that's actually a perfectly accurate bug report. Musepack seeking doesn't work at all...
12:54:31linuxstbpreglow disabled it a while ago.
12:54:40B4gderah, I misread it
12:54:51B4gderI missed the "seeking" part
12:55:15*B4gder shuts up
12:55:36Rob2222XavierGr: Yes, the power connector.
12:56:02preglownice bug report
12:56:15XavierGrRob good idea. But I will have to make cables come out of my case.
12:56:33preglowanyone going to respond, or shall i?
12:58:14preglowok, i'll do it
13:03:38NightCatCould someone explain me my problem...
13:03:56XavierGrwhat's your problem
13:04:06NightCatWhy only unifont and some fonts named like 6X9 works on my DAP?
13:04:21***Saving seen data "./dancer.seen"
13:04:27XavierGrwhat dAP?
13:04:46NightCatH320 I've just tried all that fonts:">
13:04:57NightCatBut non is working...
13:05:02B4gderNightCat: because you have no other unicode font
13:05:14NightCatI mean they works but not show russian symbols...
13:05:28NightCatThen what shall I do?
13:05:29amiconnThese fonts are from before uncide support in rockbox
13:05:37amiconn=>they won't work anymore
13:05:59NightCatThen if I want some more fonts I should convert it myself?
13:06:37NightCatThen when can I get bdf to RB font program?
13:06:47 Quit JdGordon (Read error: 104 (Connection reset by peer))
13:06:49B4gderNightCat: in the rockbox source code
13:07:13NightCatYou mean in tools?
13:07:54*B4gder looks at NightCat
13:08:06B4gderdo you _ever_ check anything for yourself instead of asking?
13:08:43NightCatI like asking and talking :)
13:09:07NightCatBut when I ask something I start to find it myself :)
13:09:08B4gderthen let's hope someone likes answering
13:10:06NightCatYes :)
13:10:11NightCatLet's hope :)
13:11:04Kingstonecan anyone give me link to rockbox sourcecode for ipod?
13:11:17B4gderKingstone: it is the one and single source for all targets
13:11:40NightCatKingstone: Just use the CVS.
13:11:42Kingstoneso can you give me link? or to he cvs?
13:11:59B4gderI'm sure you find it
13:12:28NightCatSomething like that I think...
13:12:57LinusNwhat's with all lazy people today?
13:13:25B4gderone of these days I'll go join a #rockboxdev channel instead
13:13:39*B4gder is grumpy
13:13:46*linuxstb is sleepy
13:15:25NightCatB4gder: We'll find you anyway :)
13:15:26LinusNi wonder how one could invent an audio format that doesn't support seeking
13:15:35preglowsheer cleverness
13:15:57linuxstbNot many audio formats do have good seeking support.
13:16:00preglowB4gder: i already have #rockbox-dev :-)
13:16:13B4gdernice and quiet there, huh? ;-)
13:16:20Rob2222LinusN: Maybe for some special cases where you didnt need to seek
13:17:12preglowplenty of action!
13:17:28preglowwe have free beer
13:18:08preglowas long as you have it, that is
13:18:28*stamppot trudges over to #rockbox-dev.... :)
13:19:01stamppotWhoa! Loads of free beer! :)
13:19:57preglowLinusN: they managed it by making sure that all parts of a stream might depend on any other previous part
13:20:08XavierGrwhat? rockbox-dev?
13:20:12 Join muesli__ [0] (n=muesli_t@
13:20:19preglowLinusN: some kind of infinite duration bitreservoir or something
13:20:33B4gderand please head to #rockbox-questions when you need to ask silly questions
13:20:50B4gderand #rockbox-spam if you feel a need to spam
13:20:56damakino, it is much more fun to annoy people here :P
13:21:26stamppot#rockbox-spam is now open for submissions...
13:21:31NightCatMaybe #rockbox-fun or something?
13:21:38Rob2222#rockbox-offer <= today 50% cheaper rockbox sales
13:23:49Kingstonewhat is outl()?
13:24:06linuxstbKingstone: A linux-ism that needs to be removed.
13:24:18B4gderKingstone: linux lingo for writing a 'long' to memory
13:24:30 Join Cassandra [0] (
13:24:55Kingstoneok thanks
13:25:04preglowlinux usually runs in user mode, so you can't just write to any address just like that
13:25:08preglowso they wrap it
13:25:27B4gderwell that's actually a macro for drivers
13:25:35B4gderand they're not user-space
13:25:42CassandraHmmm. What's a "data abort"? That's a new one on me.
13:25:49preglowCassandra: you just tried to access unaligned memory
13:26:08preglownote the address and fine out where for bonus points
13:26:21CassandraThis is a colour wps.
13:26:22preglowB4gder: oh?
13:26:26preglowB4gder: what does the macro do?
13:26:29CassandraNot sure why it would be doing that.
13:26:32linuxstbIt's possible the bitmap loader is at fault.
13:26:40B4gderpreglow: on uclinux I don't think it does anything magic
13:26:55B4gderpreglow: but I think it can do memory remapping thing
13:26:59Cassandralinuxstb, I had a look at it last night. It's ... opaque.
13:27:17B4gderI'll admit I'm not a wizard on that macro
13:27:18CassandraI think we also need to increase the image buffer size as a function of display depth.
13:27:36B4gderbut I won't admit _anything_ else
13:27:40CassandraBut not understanding the bitmap loader, I wasn't able to work out what that function should be.
13:28:37LinusNCassandra: the wps image buffer is not in the bitmap loader
13:29:09CassandraNo, but it needs to have enough room to contain al the images loaded.
13:29:36preglowB4gder: so, dealing with linux, are you?
13:29:41LinusNCassandra: yes
13:29:49linuxstbYes, it seems the readlong function needs to do unaligned memory accesses on little-endian targets.
13:29:57*linuxstb fix
13:30:01CassandraHmm. The bitmaps causing the problem are 9x13 and 11x11 btw.
13:30:03B4gderpreglow: doing the occational device drivers hacking at work
13:30:54linuxstbIn fact, I think the readlong and readshort macros should just be used for all targets - big or little endian.
13:31:49LinusNCassandra: which target and which wps?
13:31:51linuxstbCassandra: Can you try removing the #ifdef ROCKBOX_LITTLE_ENDIAN from line 71 of apps/recorder/bmp.c so that those readlong and readshort functions are used on all targets, not just big-endian targets?
13:32:03XavierGrThe other day I found a strange button combo on H140
13:32:21Rob2222Jejeje, my H300 harddrive update just arrived!!! :)
13:32:44Cassandralinuxstb: Sure. Is that a wise thing to do though?
13:32:45Rob2222 <= I only doesn't know how to put it in my H300, yet.
13:32:49XavierGrIf you press play + a-b in while playing screen it will go to ID3 tag information!
13:33:01linuxstbCassandra: Yes, I think so.
13:33:11XavierGrRob you already had a 60GB drive so...
13:33:30Rob2222XavierGr: Yes, but that are 250GB :D
13:33:31CassandraI think gwps.h line 275 needs changing too.
13:34:21XavierGrRob I thought you said H300 HD update.
13:34:43linuxstbCassandra: That would make sense to me. I don't understand the logic behind that formula though.
13:35:06linuxstbAFAIK, it hasn't been a problem for H300 users.
13:35:43B4gderlinuxstb: for the sim perhaps?
13:35:55CassandraThere used to be an 8 in there.
13:36:01CassandraI wonder where it went.
13:36:18linuxstbB4gder: There's no point IMHO. It's not a time-critical function, so we may as well just read the values byte-by-byte in all cases.
13:36:26Rob2222XavierGr: As soon as I found a way to fit in in my H300. ;)
13:36:30B4gderI agree
13:36:30Rob2222OK, was just kidding.
13:36:43*Rob2222 shuts up.
13:37:29preglowperhaps the codec api configure function should have a CAN_SEEK thing
13:37:58linuxstbOr the codec should reply "can not seek" when the playback code requests a seek.
13:38:10linuxstbThis will take care of codecs like FLAC where certain files are not seekable.
13:38:38linuxstbSorry, I've misunderstood - the codec will set CAN_SEEK for each file after it opens it?
13:38:56preglowanything else would be a bit unflexible
13:38:59linuxstbThat would work.
13:39:09preglowit's just that i don't want seeking to appear to work
13:39:11preglowas it does now
13:39:43linuxstbBut all codecs should be able to seek back to the start of the track.
13:40:23CassandraLinus made the buffer take up 8 bytes per pixel in 1.17 of gwps.h I think.
13:40:38CassandraIt'd work, but it seems incredibly wasteful.
13:41:12Cassandra(The divide by 10 a 'fudge factor' arrived at experimentally, I believe.)
13:41:46preglowlinuxstb: well, sure, but i don't see that as seeking
13:41:59preglowlinuxstb: i don't even think it's handled the same by the playback engine
13:42:18linuxstbThere is no "restart track" event - it's passed to the codec as "seek to 0"
13:42:37linuxstbBut it's probably not an issue...
13:43:22Rob2222H300: 1.30: Released December 2005 This release is for the European firmware only. Due to licensing issues in Europe, it removed the video icon & mode added in release 1.29.
13:43:59NightCatRob2222: Don't you know why?
13:44:39CassandraLinusN, sorry - missed your question. It's a colour version of boxes and iPod platform.
13:44:49Bg3rRob2222 this is f.cking bullsh.t ... <sorry for the language>
13:44:53Rob2222Bg3r asked why 1.30 has removed video support.
13:45:06Rob2222So thats wrong?
13:45:12muesli__NightCat it has probably something to do with a certain zax on multimedia players
13:45:46muesli__at least its a guess
13:45:56Rob2222muesli__: Have you tried 1.30 Eu?
13:46:15muesli__never used irivers firmware so far ;)
13:46:35NightCatWindows'll now be without Windows Media Player (EU version)... :)
13:46:36 Join webguest57 [0] (
13:46:54*muesli__ is absolutely rbx-fanatic ;)
13:47:14Rob2222well, EU version should have lower sound outpout, only due to french regulations, too.
13:47:39NightCatF*cking EU lows...
13:47:43Rob2222I didnt think that it is f.cking bullsh.t
13:47:55muesli__is the korean version louder in fact?
13:47:58Rob2222NightCat: Simply use the KR version :)
13:48:07Rob2222muesli__: I didnt have tested it, yet.
13:48:09NightCatmuesli__: Yes.
13:48:26NightCatI think Yes :)
13:48:27muesli__i patched the kr..but as i said..never tried it ;)
13:49:17NightCatWhen I patch my old imp-550 to kr it was louder...
13:49:36Rob2222muesli__: Dont listen to the german police with your iriver fm radio. ^^
13:50:00muesli__well..i rather should ;)
13:50:08NightCatHow to listen the police?
13:50:33muesli__just use the kr-version providing lower bands on radio
13:50:33 Quit webguest67 ("CGI:IRC (EOF)")
13:50:53NightCatOnly 87 or something... NO?
13:51:22muesli__afaik it was something like 72mhz
13:51:44NightCatNow I have an kr but only 87....
13:52:01XavierGrwow the battery pack is neat!
13:52:25Rob2222NightCat: You have to set the tuner region.
13:52:28XavierGralso the H340 is more fat than the H140 but more short too.
13:52:56NightCatI set t to Korea...
13:53:01Rob2222NightCat: In options. But I dont know which is the right, At least NOT usa and NOT europe
13:53:07Rob2222NightCat: Try Japan
13:53:30muesli__XavierGr the h340 is really sucking fat
13:53:36muesli__that why i didnt chose it
13:53:41NightCat76 Mhz :)
13:53:42muesli__just another brick in the wall :p
13:53:47Rob2222The FM band in Japan is 76-90MHz
13:57:45Cassandralinuxstb, still seems to be crashing, Im afraid.
13:58:20linuxstbCan you note down the address, and then look in build/apps/ to see if you can identify the function it is crashing in?
13:59:41 Quit NightCat ()
14:00:48CassandraThis is going to be a great shock to you. Looks ike it's somewhere in read_bmp
14:01:11Cassandra(00020E9C to be exact)
14:01:37 Quit linuxstb (Read error: 113 (No route to host))
14:01:43 Join linuxstb [0] (
14:02:00CassandraThis is going to be a great shock to you. Looks ike it's somewhere in read_bmp
14:02:01Cassandra(00020E9C to be exact)
14:02:07linuxstbDidn't take long:
14:03:04 Quit damaki (Read error: 110 (Connection timed out))
14:03:16linuxstbI've prepared my reply though - use the ipodlinux bootloader which has a nice menu and configuration options.
14:03:54 Join damaki [0] (
14:04:12CassandraOh, btw it's probably not a great idea for the bootloader to report a Linux error when force booting into Apple firmware.
14:05:13ashridahwhy not? people blame linux for all sorts of things that has nothing to do with linux itself :)
14:06:32CassandraOur keyboard driver needs to do clicky clicky like Apple.
14:06:44CassandraClicky clicky is a must have feature.
14:06:50linuxstbCassandra: When does it do that? It should only do that if you press PLAY (to start Linux). MENU will take you to the Apple firmware without any errors (I hope)
14:07:49CassandraYou are right, yes. I'd misremembered the boot key.
14:08:54CassandraIt would be really nice if I could find a utility that would generate me an ipod database in the same way we generated the tagdb.
14:09:13CassandraI don't want my MP3s managed. I just want iPod to realise they're there.
14:09:17linuxstbJust wait for tagcache...
14:09:46preglowCassandra: gnupod?
14:10:09CassandraDid I mention I wanted it to run under Windows?
14:10:18preglowi think gnupod is perl
14:10:58preglowthere's also a foobar2k component
14:11:54Cassandrathe foobar2k one looks like it copies MP3s about too. Most of them seem to assume you want some sort of itunes like functionalty on the host.
14:12:21 Quit webguest92 ("CGI:IRC (EOF)")
14:16:18*linuxstb sees a Rockbox plugin being developed...
14:16:40CassandraPreferably by someone other than me.
14:18:07CassandraSo, any more bright ideas on this unaligned access?
14:18:52preglowdid you found out where it occured?
14:19:00preglowoh, read_bmp
14:19:07*Cassandra nods.
14:19:26CassandraContain your shock if you can.
14:19:49webguest57heh imagine a rockbox plugin that does the itunes database thingy, ie acts as ums but puts files in these weird directories and updates the database on umount
14:20:26linuxstbCassandra: Do you mean read_bmp_file()
14:20:45damakiwebguest57: it would be damn slow. reading all the tags would be soooo long
14:20:47CassandraYeah, sorry.
14:21:28CassandraIm in insomniac mode at the moment thus not terribly with it.
14:21:51preglowdamaki: not _that_ slow, the file data would have to be read anyway
14:22:37linuxstbCassandra: I can't see anything in that function that is obviously doing unaligned accesses - apart from the readlong and readshort, which should now be fine.
14:22:40 Join damaki_ [0] (
14:23:53CassandraYou wanted the version of readlong and readshort that did all the byte munging, yes.
14:24:14linuxstbYes. But I think I've found the problem now.
14:25:39 Quit Matze (Read error: 110 (Connection timed out))
14:26:46preglowand i think i've found a new scratch register!
14:27:46preglowyeah, arm seems to have five scratch regs
14:27:51preglowr0, r1, r2, r3 and _r12_
14:27:53linuxstbCassandra: (one line)
14:27:57 Join kernel_sensei [0] (n=boris@
14:28:26linuxstbCassandra: Ignore that patch for now, just found a bug....
14:28:40 Quit damaki_ ("You're entering a world of pain")
14:29:06linuxstbCassandra: OK, you can try it now.
14:29:34*linuxstb welcome r12 to the party.
14:30:49Papricahow do i need to use the read_bmp_file function?
14:32:00LinusNPaprica: what do you want to know?
14:32:08 Quit kernelsensei (Connection timed out)
14:32:10Papricamm i try to use it
14:32:15 Nick kernel_sensei is now known as kernelsensei (n=boris@
14:32:30Papricabut the simulator stuck up
14:32:47LinusNPaprica: gdb is your friend
14:32:52Papricais this right?
14:32:52Papricaread_bmp_file("/.rockbox/icons.bmp", &bm,
14:33:04Papricamm i cant compile it =\
14:33:06Papricadont know why
14:33:25 Join tvelocity [0] (
14:33:48LinusNwhat does the compiler say?
14:34:06Papricamm sec i compile it again now
14:35:00Jungti1234hi Paprica
14:35:42 Quit linuxstb (Read error: 104 (Connection reset by peer))
14:35:48 Join linuxstb [0] (
14:36:13LinusNPaprica: why are you building the gdb stub?
14:36:29Papricadont understand
14:36:41LinusNhow did you configure?
14:37:10LinusNyou are building the archos gdb stub
14:37:26Papricamm so what i need to do
14:37:45LinusNyou need to reconfigure
14:37:55Papricaand what to type
14:37:57LinusNyou have obviously selected the wrong target
14:38:09Paprica10 is h300...
14:38:24Paprica10 - iriver H320/H340
14:38:28 Quit damaki (Read error: 110 (Connection timed out))
14:38:38preglowyou're still building the gdb stub
14:38:39preglowwhy 'g'?
14:38:40linuxstbAnd then you just want "N" for normal, or "S" for simulator - not G.
14:38:55LinusNPaprica: you have probably selected "1" instead of "10"
14:39:20linuxstbLinusN: There is a "G" option now for all targets...
14:39:31CassandraBuildy build build
14:39:33preglowi still think this should be fix
14:39:39B4gderme too
14:39:43linuxstbme too
14:40:14Papricaso how do i compile the gdb for h3xx?
14:40:23B4gderbuild the sim
14:40:25B4gderrun gdb
14:40:39LinusNyou don't compile a gdb stub for h300, there is none
14:40:42CassandraOh btw. Don't buy Sony. I've just had to scrap my 1800 pound laptop I bought 2 years ago because it fell apart.
14:40:58Papricamm where i can find a gdb for windows?
14:41:13B4gderPaprica: install a proper cygwin and it'll be there
14:41:19Rob2222Cassandra: Yes, sony laptop has a really bad warranty service when over the 1 / 2 years
14:41:28LinusNPaprica: by installing a proper cygwin installation and not tha silly devkit
14:41:40Papricai have the cygwin
14:42:01B4gderthen you just re-run cygwins install thing and select gdb in there
14:42:19LinusNjust like the installation instructions told you to
14:42:26CassandraLinusN, btw why did you multiply the image buffer size by 8? 24 bit colour is still only 4 bytes/pixel.
14:42:41LinusNCassandra: where?
14:43:30CassandraWhen adding colour support.
14:44:12LinusNwhere in gwps,h?
14:44:40 Quit B4gder ("time to say moo")
14:45:55LinusNah, now i see
14:45:59Cassandralinuxstb, nice work. That's got it/
14:46:06 Quit Jungti1234 ()
14:46:09CassandraWant me to commit it for you?
14:46:09LinusNthat has nothing to do with 4 bytes per pixel
14:46:27LinusNas you know, the internal format is 2 bytes per pixel
14:46:32linuxstbCassandra: Yes, please go ahead.
14:47:09linuxstbCassandra: Before you do that, can you put #if LCD_DEPTH == 16 around that change I made to increment ret.
14:47:34LinusNhowever, the images are in fact 16 times bigger than the monochrome ones in full color
14:47:40linuxstbCassandra: There's no point doing it for targets with char* framebuffers.
14:48:09CassandraAre they? How come?
14:48:33linuxstbMono is 1bpp, Color is 16bpp.
14:48:39LinusNCassandra: easy: monochrome == 1bit per pixel, color == 16bits per pixel
14:49:21*Cassandra nods. So why'd you only multiply by 8 then?
14:49:29LinusNbecause i didn't care
14:49:40preglowlinuxstb: this ui sluggishness is probably going to promote the task of making the cop work
14:49:42LinusNthe buffer size is pure guesswork anyway
14:49:43CassandraSorry if I'm being slow here.
14:50:06*Cassandra nods. True, although I've had to increase it in the past.
14:50:07linuxstbpreglow: Maybe we should try other things first - like making an asm version of the fiq.
14:50:16preglowi can do that
14:50:24LinusNi just removed the /8 to make it bigger
14:50:24linuxstbIt would be great if we could make Rockbox work with only one cpu.
14:50:37preglowlinuxstb: well, that'll primarily mean optimising codecs, then
14:51:14preglowdumped saving of r12 in thread context now, and everything works fine still
14:51:21preglowwait a bit, and i'll commit it
14:51:41linuxstbIt's possibly best to wait before touching the fiq routine first. We should iron out any bugs in the pcm playback code first.
14:52:14preglowdisk cache works on nano
14:52:18preglowgod knows what good it does us
14:52:25linuxstbAny noticable difference
14:52:37preglowi just had a data abort
14:52:44preglowin a codec
14:53:02linuxstbI think I've had one - but I forget which codec. Possibly wavpack.
14:53:31 Join Cari [0] (n=zeroirc@
14:54:01preglowit was either wav or wavpack
14:54:09preglownot reproducible
14:55:27preglowpossible wavpack, the function the address maps to is read_next_header
14:55:40 Quit Cari (Client Quit)
14:59:00preglowme goes brb
15:02:30 Join damaki [0] (
15:04:25***Saving seen data "./dancer.seen"
15:06:54linuxstbpreglow: I think it's the line "* (long*) cp = temp" in little_endian_to_native() in libwavpack/bits.c
15:06:56linuxstbThe reverse function has a similar bug.
15:07:38 Join ]RowaN[ [0] (
15:07:57]RowaN[guys all of a sudden my wps bitmaps seem to be inverted.. have i missed something?
15:10:50linuxstb]RowaN[: Which player are you using?
15:12:39 Join Genre9mp3 [0] (
15:19:52]RowaN[its been like that using builds from about 1 day ago and onwards
15:20:05 Join Kohlrabi [0] (
15:20:09linuxstbBy "inverted", do you mean that black is white, and white is black?
15:20:33]RowaN[i know theres a simple solution.. invert my bitmaps =]
15:21:03]RowaN[but has it happened to anyone else?
15:22:11muesli__ive experienced that issue too
15:22:19muesli__did u use photoshop?
15:23:35 Join tucoz [0] (
15:23:48tucozThe ipodVOL theme seem to have the same problem
15:24:26tucozwith the next to most recent bleeding edge
15:25:54 Quit ashridah ("Leaving")
15:26:16muesli__"Small ARM opt. Saves eight bytes!" is that sarcasm or real progress`?
15:27:56]RowaN[photoshop yes
15:28:49muesli__dunno what goes wrong at this point but inverting helps
15:29:46]RowaN[since day 1 my wps bpms have been inverted (thats how i had to make them in order for them to display non-inverted in rockbox) ... but only in the last fews days they actually show up as they are
15:29:52]RowaN[bmps even
15:30:17 Part tucoz
15:30:27]RowaN[which way are they supposed to be saved? normal or inverted?
15:30:39lostlogicpreglow: in case you didn't see, I posted a newer version of my quick codec timing patch to sourceforge that gives reproduceable results and stuff... still hacky, but should give you something to measure progress with while optimizing on arm.
15:33:25preglowlostlogic: i'm sure it's more than adequate
15:33:31preglowwill test it when i'm not bogged down with work :/
15:33:49lostlogicpreglow: nod nod enjoy ;)
15:34:10 Join linuxstb_ [0] (
15:34:18lostlogicpreglow: I hope when I'm done with hacking deep into the bowels of the audio system I'll be able to make a better version that is CVS worthy.
15:35:16linuxstb_lostlogic: Have you looked at the ipod audio code at all? (in pcm_playback.c)
15:35:24]RowaN[grr i just inverted my bitmap and saved it... wps STILL shows colours the wrong way around
15:35:42 Quit linuxstb (Nick collision from services.)
15:35:43 Nick linuxstb_ is now known as linuxstb (
15:36:32lostlogiclinuxstb: slightly −− enough that I didn't break playback when I modified the callback structure in the pcmbuffer
15:36:55preglowlostlogic: i'm looking forward to seeing the result
15:37:30linuxstblostlogic: I was wondering if I was doing anything stupid in it - I don't understand the playback system...
15:38:19lostlogiclinuxstb: nothing stuck out as stupid, but I'll give it another look at some point I'm sure
15:39:19linuxstbI think I may have found a bug anyway - the "unpause" event doesn't seem to be handled by my code. So I think that's why skipping to the next track stops the playback.
15:40:49CassandraSoon, soon it will be time to wind up amicon.
15:40:55 Join Maxime [0] (
15:41:03lostlogiclinuxstb: track skips are weird beasts, they involve at least one call to pcm_play_stop, pcm_mute and pcm_mute(false) don't think they involve pause and unpause though
15:42:10linuxstbWhen is pcm_play_data() called?
15:42:23lostlogicwhen the first sample of the new track is decoded
15:42:40Cassandra(Possibly by spelling his name wrong, but more likely by commiting another wps.)
15:42:43lostlogicOR on a skip on disk after the new track is buffered and several sammples have been decoded
15:43:05linuxstbSo it should be called when a user skips to the next track?
15:43:20linuxstbCassandra: Is that a colour WPS for the Nano?
15:44:04CassandraHmm - I don't think I meant to change button.h. How does that get reverted?
15:44:20CassandraOr how do you revert to the previous CVS version?
15:46:31Bg3rwith commiting it again?
15:46:59 Join XavierGr_ [0] (
15:48:26CassandraEasiest way, I gues.
15:48:55 Join linuxstb_ [0] (
15:49:19 Quit linuxstb (Nick collision from services.)
15:49:22 Nick linuxstb_ is now known as linuxstb (
15:49:38*linuxstb is having a bad ADSL day...
15:49:55preglowevery day is a bad adsl day
15:50:40 Join godzirra_ [0] (
15:51:15 Quit DangerousDan (Read error: 104 (Connection reset by peer))
15:51:33linuxstbCassandra: Do something like "cvs diff -r 1.34 button.h | patch -R" and then commit
15:52:00 Join ts-x [0] (
15:53:51ts-xSlasheri: You may already be aware of this, but I think there is a bug in the wps cache code. If you connect via usb, modify and existing wps image, and reload the wps the old image still appears in the wps.
15:54:33preglowand that's exactly what i warned him would happen
15:54:38preglowi think it's time revert the bmp cache
15:54:45ts-xThe only way I've found to get around it, is by changing the wps name while in usb and loading the new wps.
15:55:07 Join DangerousDan [0] (
15:55:24linuxstbts-x: As a workaround, could you load another wps, and then go back to your wps?
15:56:23linuxstbpreglow: A revert would get my vote (sorry Slasheri...)
15:56:32preglowi just think it was a bad idea in the first place
15:56:37ts-xlinuxstb: I don't believe that will work, but let me test...
15:57:26 Quit godzirra (Read error: 110 (Connection timed out))
15:57:47BHSPitLappyis better color customization in sight in a cvs scenario?
15:57:53CassandraRight - iPod Rockbox is now officially in glorious colour.
15:58:01preglowthere's no way to verify the cache is correct, since doing so would mean opening and scanning all those files anyway
15:58:54preglowlinuxstb: looks like the wolfson chip is going to give us tons of fun when reclocking the cpu
15:59:15preglowi wish i had the slightest idea how this worked...
15:59:51lostlogicmaybe that's why IPL uses one processor for the data transfer to the DAC so that it can keep that processor constantly clocked low rather than reclocking?
16:00:07linuxstbNo, it doesn't do any reclocking at all.
16:00:21linuxstbAnd you can't independently clock the two cpus.
16:00:27*Cassandra is particularly proud of the volume control. (Oh gimp, how I adore thee.)
16:00:44lostlogichmph ah.
16:01:17linuxstbCassandra: My ipod's got a 220x176 screen.... :( But I'm about to give it a look.
16:01:50*preglow checks out
16:01:59CassandraNo facilities here, sorry.
16:02:37Cassandra(At the moment I'm looking to go to a 60gb iPod when I fill my 40gb drive.)
16:02:51 Join DJ_TCK [0] (
16:03:03 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
16:03:08Cassandra(Which I estimae being in 3-6 months time. I hate not having *all* my music with me.)
16:03:24ts-xlinuxstb: Loading another wps and then reloading the existing is kind of inconsistent. Sometimes it corrects, sometimes it doesn't. What's also strange is that last night none of the image changes I was making seemed to take, but this morning the first image change did update while the second did not.
16:04:12CassandraI tried taking one with my phone, webguest57, but iPod nano screen + crappy phone camera meant I couldn't get a nice picture.
16:04:19preglowi get a data abort
16:04:22preglowperhaps cvs up
16:04:23ts-xWhether or not the image was directly modified on the player or modified on the pc and copied over seems to have an impact as well.
16:04:54Cassandrapreglow, yes. Needs the current bleeding edge I'm afraid.
16:05:53preglowCassandra: nicety
16:05:59Slasherits-x: yes, true. That problem is not possible to solve in easy way with the current cache
16:06:05*linuxstb wants the DancePuffDuo in colour...
16:06:23Slasherii think the real solution would be the combined bitmaps or early dircache (i am working with that)
16:06:24CassandraAh, I do not the silly gimicks.
16:07:51ts-xSlasheri: Not a bid deal, just wanted to make sure you were aware. Just made me think I was losing it for a few minutes last nght until I figured out what was going on :)
16:08:15 Quit XavierGr (Read error: 110 (Connection timed out))
16:08:27 Join IzySad [0] (
16:08:29preglowSlasheri: i think it should be reverted until the issue is fixed
16:08:33preglowSlasheri: it's confusing
16:09:23CassandraYou know, we could really do with a rockbox@home distributed build system. Anyone fancy writing one?
16:10:33SlasheriCassandra: Hmm, do you mean with that?
16:10:37Slasherisounds like an interesting idea
16:10:50Slasheripreglow: Hmm, maybe true
16:11:05SlasheriCassandra: +what
16:11:21IzySadhey, i was thinking about modding my creative nomad zen. i've never programmed at the firmware level. would it be worth looking at rockbox code or is it all going to be different? any tips or how to get started?
16:11:35Slasheriit is different..
16:11:35preglowSlasheri: well, i know i would confused if i tried to update my wps and nothing happened
16:12:14linuxstbIzySad: After you work out how to compile and run your own code on the zen, and work out how to access all the hardware, then you can port Rockbox to it. But up to then, Rockbox probably isn't much help.
16:12:15IzySadany tips on how to start since it's different? how did rockbox go about starting?
16:12:19Slasheripreglow: yes, true.. but without more complex caching on background, i don't think that is fixable
16:12:30preglowSlasheri: i don't think it's fixable either
16:13:01SlasheriIzySad: yep, the first thing would be to make your own code running on the device before even thinking about portin rockbox..
16:13:10linuxstbIzySad: Disassembling a player, finding datasheets for all the chips, unsoldering chips from the PCB to draw schematics....
16:13:12preglowi vote for removing it, feel free to see what the others say
16:13:33IzySadwow, so it would have to go to the hardware reverse-engineering level?
16:13:33Slasheripreglow: ok, i don't mind removing that.. :)
16:13:52SlasheriIzySad: absolutely
16:14:06beebleIzySad: if you didnt get any zen sourcecode, yes
16:14:07preglowsome other scheme to do the same thing that works 100% is welcome
16:14:19linuxstbIzySad: Or the alternative is to reverse-engineer the Creative firmware.
16:14:29SlasheriIzySad: probably you even need to desolder all components and destroy few players in order to trace the pcb
16:14:38linuxstbMost Rockbox ports involve a little of both, but mainly hardware disassembly.
16:14:53IzySadlinuxstb: right, i was thinking reverse engineering the firmware might be a reasonable way
16:15:05]RowaN[so guys is there any way to fix my inverted bitmap problem?
16:15:22linuxstbIzySad: The ipodlinux people have never opened an ipod afaik.
16:15:33Slasheripreglow: i think early dircache should do that and even speed rockbox loading in general as it affects to all open()..
16:15:42CassandraSimple really - Rockbox builds currently take about half an hour. The reason is that the machine has to build about 15 different Rockbox targets. If you could pass off compilation to volunteer machines, you could instantly parallelise the cross platform builds Reducing the time drastically.
16:16:12IzySadlinuxstb: thanks, that last comment gives me some hope, i'll go check out that code...
16:16:14 Quit IzySad ()
16:16:23CassandraAlso it'd be nice if developers had access to a pool of machines so they could check code compiles for all architectures before commiting.
16:16:29preglowSlasheri: how will early dircache work?
16:16:30 Join frederic_ [0] (
16:16:47Slasheripreglow: it loads the cache from disk (when it's not marked as dirty)
16:17:04preglowSlasheri: isn't that the feature the other devs said no to when you first commited the dircache?
16:17:15preglowafaik, you had this feature some time ago already
16:17:21Slasheripreglow: yes, but with minor changes that is possible to do easily
16:17:27beeblelinuxstb: there are other sources to get part numbers. without them its very silly just to reverse eng. the firmware imho
16:17:29preglowok, sounds fine to me
16:17:35Slasherii will try to integrate that into the next bootloader including flash support
16:18:05linuxstbbeeble: Of course. But someone has to disassemble a player to get the part numbers.
16:18:15 Quit Cassandra (" Want to be different? HydraIRC -> <-")
16:18:24linuxstbI was just saying how Rockbox ports have happened in the past.
16:19:19beeblelinuxstb: i was jsut refering to the ipodlinux guys didnt open any ipod
16:20:10linuxstbThat's true. They got the chip information from elsewhere - but more information on how the chips are wired together would help.
16:20:19beebleapropos open ipod. i must find a source for 4gb samsung nor. does anyone have a broken nano?
16:26:09 Quit DJ_TCK (Read error: 110 (Connection timed out))
16:27:53Bg3ramiconn you there ?
16:29:27 Nick XavierGr_ is now known as XavierGr (
16:30:09 Quit CoCoLUS (Read error: 110 (Connection timed out))
16:32:07 Quit billytwowilly ("Leaving")
16:33:42 Join bluey [0] (
16:34:22 Quit ts-x ("CGI:IRC (EOF)")
16:34:50 Join DT291 [0] (
16:37:00 Part XavierGr
16:37:53 Quit ]RowaN[ ("CGI:IRC (EOF)")
16:39:18 Quit Genre9mp3 ()
16:42:24 Join CoCoLUS [0] (
16:45:09 Quit DreamTactix291 (Read error: 110 (Connection timed out))
16:52:58 Join DangerousDan [0] (
17:03:30 Join LuMiNoU [0] (
17:04:26 Join zhilik [0] (
17:04:30***Saving seen data "./dancer.seen"
17:04:45Rob2222simple colored text isnt possible in wps, or?
17:05:25zhilikDoes anybody know how the iAudio X5 port is going?
17:15:19LuMiNoUHi there :) Is there a way to completly remove the RockBox bootloader on an iPod Color ? in order to re-use the original bootloader ? (of course if i can't loose all music it will be better :p)
17:16:13linuxstbLuMiNoU: Do you still have the bootpartition.bin file you created during the installation?
17:16:25 Quit DT291 (Connection timed out)
17:16:33linuxstbThen you can just write that back to the disk.
17:16:44linuxstb... using ipodpatcher
17:16:54LuMiNoUipod_fw -w bootpartition.bin ?
17:17:05 Quit Zagor ("Client exiting")
17:17:14linuxstbWith the disk number - the same command you used to write rockboot.bin
17:17:45LuMiNoUOkay, trying it
17:17:50linuxstbi.e. ipodpatcher -w N bootpartition.bin
17:18:31LuMiNoUYeah, works perfectly
17:18:34LuMiNoUThx :)
17:19:23Rob2222The apple music database is not overwritten this way?
17:19:35linuxstbRockbox goes nowhere near the itunesdb
17:19:49linuxstbIt's just stored in a file on your FAT32 partition - under iPod_Control
17:19:57Rob2222ah ok
17:20:13zhilikI've scanned the iAudio X5's PCBs:
17:21:21 Quit LuMiNoU ("Thx guys :)")
17:21:51 Join Matze [0] (
17:22:06 Quit tvelocity ("Αναχωρεί")
17:23:15 Join gtkspert_ [0] (n=gtkspert@
17:24:30 Quit bluey ("Leaving")
17:27:57 Join bluey [0] (
17:34:20 Quit frederic_ ("Chatzilla [Firefox 1.5/2005111116]")
17:35:12 Quit gtkspert (Read error: 110 (Connection timed out))
17:36:25 Quit Matze (Nick collision from services.)
17:40:59 Join Matze41 [0] (
17:55:03 Join safetydan [0] (
17:58:03safetydanAre codecs loaded in a different manner than plugins?
17:58:32lostlogicthey are quite similar, but go into different areas of RAM
17:58:35linuxstbThey are loaded slightly differently, but the principle is the same.
17:58:51lostlogiclinuxstb: love the relatively easy questions :)
17:59:12linuxstbYes, if there are any follow-up questions, I'll be lost....
17:59:14safetydanI'm trying to narrow down why plugins load in the Win32 SDL sim, but not codecs
17:59:22lostlogicsee, and then we're lost :-D
17:59:25linuxstbPlugins and codecs work differently in the sim.
17:59:31linuxstbcompared to the real targets
17:59:40safetydanIs there any reason for that?
18:00:11linuxstbYes - on the target we are dealing with binary images loaded to fix areas in RAM. In Windows, we use dlls
18:05:19safetydanExcept plugins aren't DLLs in Windows and they load fine
18:05:35preglowperhaps i should try the iram remap we were talkinga bout now
18:05:56 Join Zoide [0] (n=aodio@aodio.Stanford.EDU)
18:06:00 Nick Zoide is now known as Zoide777 (n=aodio@aodio.Stanford.EDU)
18:06:25Zoide777is it possible to install Rockbox on a B&W display 4G?
18:07:57preglownot yet
18:07:57 Quit Zoide777 (Read error: 104 (Connection reset by peer))
18:11:38 Join linuxstb_ [0] (
18:13:02 Part safetydan ("Leaving")
18:14:23 Join BHSPitLappy2 [0] (n=Steve-O@
18:14:43linuxstb_preglow: An IRAM remap would be very nice.
18:14:55linuxstb_Did you see the problem with the Color LCD?
18:15:34preglowunless something new has been uncovered
18:15:51amiconnBg3r: Now I am...
18:15:52linuxstb_I mean did you see how I fixed it?
18:16:43linuxstb_The ipod hardware revision (e.g. 0x60000) is read from memory location 0x00002084 - this obviously didn't work when you remapped SDRAM over the top of that memory location.
18:16:56preglowriiight, of course not
18:17:29preglowbah, i'd rather not do this remapping
18:17:32linuxstb_So the sort-of hack that I did was to read that value in the bootloader, save it in the last 4 bytes of SDRAM, and then read it back at the start of Rockbox.
18:17:42preglowit messes up rolo
18:17:54linuxstb_How do you mean?
18:17:55preglowwe need to preserve that value if we want rolo to work
18:18:07preglowright now it might be overwritten by the audio buffer, i take it
18:18:15 Quit linuxstb (Read error: 110 (Connection timed out))
18:18:24linuxstb_No, it's saved into a normal variable immediately Rockbox starts (in system_init)
18:18:41preglowok, so we'll have to write it to the end of ram again in rolo?
18:19:25linuxstb_But I'm not sure rolo is worthwhile anyway. USB disk mode always reboots anyway.
18:23:35preglowwell, perhaps not now'
18:23:42preglowbut if we ever make our own usb driver
18:25:08Papricafucking cygwin
18:25:18 Quit DangerousDan (Read error: 104 (Connection reset by peer))
18:27:20 Join DangerousDan [0] (
18:27:22Papricanow its good
18:27:26Papricathanks god
18:29:46 Quit BHSPitLappy (Read error: 110 (Connection timed out))
18:33:48 Join NicoFR [0] (
18:42:03 Join damaki_ [0] (
18:42:32Papricamm after i open the gdb
18:42:42Papricawhat is the command that i need to write?
18:46:14preglowto do what?
18:46:28Papricato debug the simulator
18:46:41preglowdepends how you want to debug it
18:46:51preglowif you just want a crash log, just 'run'
18:46:57preglowand use it like normal, until it crashes
18:47:04preglowthen do 'bt' to find out where it did
18:47:12preglowfor other stuff, i recommend you find a gdb tutorial
18:47:15preglowi'm not an expert
18:48:19Papricaok thenx..
18:49:03Papricamm can you help me with read_bmp
18:50:30amiconnOh no..
18:50:40*amiconn spots another bitmap monster in cvs
18:50:42preglowPaprica: no, i can't help with anything right now
18:50:46preglowamiconn: what?
18:51:20amiconn"Colour boxes WPS for iPod (and H3xx - but it'll look a little odd on that screen. A work in progress / proof of concept."
18:51:33preglowno one is forcing you to install it
18:51:50amiconn'make zip' will package it
18:51:56amiconn(at least for H3x0)
18:52:52amiconnI should really have a look at Firefly's patch
18:53:01preglowit wont if you don't check out the wpses
18:53:09preglowwhich is a perfectly nice solution until it gets fixed
18:53:19preglowif you are going to commit that, please do ask wps makers what they think
18:53:30preglowi would not like it at all, but then again, i'm not an artist
18:53:39 Join DreamTactix291 [0] (
18:56:29 Join safetydan [0] (
18:57:01safetydanOkay, I need to stop pretending I know C. I know roughly why the codec's don't load in the Windows DSL sim, but I don't know how to fix it.
18:57:26 Quit damaki (Read error: 110 (Connection timed out))
18:58:08preglowok, why?
18:58:46safetydanWhen the SDL sim is compile, the magic DLL making stuff isn't called for the codecs
18:58:53safetydancompiled even
18:59:10SlasheriHmm, i wonder if codecs have ever worked with windows..
18:59:15safetydanMy makefile-fu is weak
18:59:18preglowthey have
18:59:20safetydanso I can't figure out how to do both
18:59:27safetydanthey do work if you compile the pure Windows sim
19:00:51safetydanSo since my (tiny) C skills aren't getting me anywhere, any way a Java programmer can help Rockbox?
19:02:57preglowtried actually coding something in c yet? :>
19:03:06preglowc isn't that hard if you do java
19:03:18 Join matsl [0] (
19:04:32***Saving seen data "./dancer.seen"
19:05:17 Quit NicoFR ()
19:05:46safetydanI can write straightforward C, it's just getting in to things like why dlopen doesn't work that I go cross-eyed
19:08:51 Join lamed [0] (
19:09:34 Join NicoFR [0] (
19:11:01preglowi wouldn't know anyway, dlopen always worked for me
19:13:57 Quit NicoFR (Remote closed the connection)
19:14:13 Join NicoFR [0] (
19:14:21amiconnsafetydan: dlopen() doesn't exist for win32, it is defined as a macro in the win32 sim
19:15:19amiconnCheck lines 296..304 in uisimulator/common/io.c
19:17:54 Join Rondom [0] (n=Rondom@
19:18:09 Nick BHSPitLappy2 is now known as BHSPitLappy (n=Steve-O@
19:18:33safetydanamiconn: I see that
19:19:02 Join slimx [0] (
19:20:41amiconnwiki spam alert :(
19:21:30safetydanI think it's a makefile thing in the end... I need to convince the codec makefile to run dlltool for the SDL sim when it's compile in windows
19:23:26 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:25:08preglowno takers for eq gui yet? :/
19:26:24 Quit linuxstb_ (Read error: 110 (Connection timed out))
19:26:48 Join linuxstb_ [0] (
19:27:50 Join Timbels [0] (
19:28:04Timbelsgood evening (netherlands)
19:30:01Timbelsanybody with some knowledge of rockbox 2.5 features awake maybe? i'm stuck with the id3 browser and can't seem to figue it out :S
19:30:41BHSPitLappypreglow: to make it?
19:31:17Timbels(i guess its very easy but i'm a big newbie) ;)
19:32:44 Quit slarti ("brb")
19:33:12TimbelsIts about the Added ID3 database support. when i go to "General Settings > File View > Show Files" and set it to "ID3 database. i see nothing (empty browser space)
19:33:27Timbelsdo i miss something here? my mp3's are nicely tagged so.. ?
19:35:02preglowhave you generated a database?
19:35:36Timbelsuhm that should be the n00by part. lol. no?
19:35:44Timbelsi assumed it would take it directly from my mp3's?
19:36:05Timbelssince the tags already inlclude genre, artist etc?
19:36:16preglowit's going to be like that
19:36:20preglowbut at the current point, no
19:36:20Timbelscan i do that inside the rockbox preglow?
19:36:25preglowand no other mp3 player works like that
19:36:27Timbelsaah ok
19:36:28preglowno, you can't
19:36:30preglowyou need a tool
19:36:36Timbelsi see
19:36:41preglowrockbox is going generate the database for you soon, though
19:36:51Timbelsthat would be wicked
19:36:52preglowjust nag at Slasheri and make him finish hit ;)
19:37:04Timbelsany clue on how that tool = called?
19:37:12preglowi have no idea, i've never used it
19:37:31preglowtry looking in the wiki on
19:37:35Timbelsdamn, but thnx for the info!
19:37:43Timbelswill do so right now
19:38:36Timbelsthat feature sounds suppurb so!! :)
19:38:47 Join RotAtoR [0] (
19:41:42 Join ssnajper [0] (
19:42:09preglowyes, it does
19:42:17preglowlets hope slasheri finishes it soon
19:42:19*preglow nudges Slasheri
19:42:42 Join damaki__ [0] (
19:43:06ssnajperi hope this isn't rude but is there any new progress on the 3g port?
19:45:30ssnajperi can't wait to see rockbox on my 3g (i donated 10$ a couple weeks ago)
19:45:50Papricastruct bitmap icons;
19:45:50Papricaread_bmp_file("/.rockbox/icons.bmp", &icons,7*144,FORMAT_ANY|FORMAT_TRANSPARENT);
19:45:56Papricawhat wrong with it???
19:46:13 Quit damaki_ (Read error: 104 (Connection reset by peer))
19:46:15 Join XavierGr [0] (
19:47:32preglowit's being worked on
19:47:33Slasheripreglow :D
19:47:37*preglow points to slimx
19:47:41preglowSlasheri: :-)
19:47:54preglowi'm dying to have tagcache :P
19:48:00Slasherihehe, i see :D
19:48:02Timbelshello id3 database sir :)
19:48:18preglowbtw, does anyone know of any other devices that support generating their own database?=
19:48:22preglowi think this might be a rockbox first
19:48:38amiconnDoesn't the iriver firmware do that?
19:48:43preglowamiconn: hell no
19:48:49preglowamiconn: you need to boot windows to do that
19:48:59amiconnI remember someone mentioning the excessive boot time with db support enabled
19:49:00Slasherihmm, that is interesting
19:49:07preglowamiconn: that's just reading it...
19:49:36Slasheriyep, with db enabled it takes 1-2 minutes to boot..
19:49:43Slasheriand it just loads it from the db file
19:49:43amiconnNah, should have known better...
19:49:50preglowwell, it sure as hell doesn't generate the db
19:49:54preglowyou need a windows application for that
19:50:03preglowwhatever it spends the time on, it's not parsing metadata
19:50:19preglowknowing iriver, they're probably just stroking the bitbucket or something
19:51:13Slasherii think what it might do is to find the corresponding pointers to the filenames from the dircache.. but the implementation has gone somehow very wrong because it's so slow
19:51:50Slasheritagcache does that also when loaded into ram but is fast with it (a few seconds only)
19:52:31Timbelshmm i miss the "genre" option Slasheri?
19:52:33preglowbut yeah, at least i know of no other devices that can make their own db
19:52:38 Join tjm [0] (
19:52:43SlasheriTimbels: yep, that is already implemented
19:52:52Slasherigenre - artist - album - songs
19:52:57Timbelsi'm trying the id3 databse mode now. looks very good! only missing the "genre" feature
19:52:57Timbelsis it? ah i used the 2.5 official build
19:53:08SlasheriTimbels: yes, i my local cvs build ;)
19:53:09Timbelsmaybe its not in there
19:53:39Timbelsso ehm, if i download the daily build should it be in there?
19:53:49Timbelsgrrrrrr :P
19:53:58Slasheritagcache is not yet there, only the old id3 db
19:54:17Timbelsok, then i will patiently wait :D
19:54:21Slasherigood :D
19:54:33 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
19:56:16XavierGrIs there any intention to port the jpeg viewer to H300?
19:56:33LinusNintention yes, effort no
19:57:36XavierGrNow that I have an H300, I am dying for it, so I will give it a try, and if I succed my filescrolling mod could see the cvs. (a very doubtfull hypothesis)
19:58:29muesli__XavierGr chaka! you can do it :)
19:59:07XavierGrWell I don't know nothing about jpeg decoding and stuff, but I will start reading and who know. The same way I started coding a little bit for rockbox.
20:00:11XavierGrthe iriver implementation of jpeg sucks big time.
20:00:31preglowahh, the h300 one
20:03:35 Quit lamed ("CGI:IRC (EOF)")
20:06:58 Join einhirn [0] (
20:07:43 Join Lear [0] (
20:13:02 Quit damaki__ (Read error: 104 (Connection reset by peer))
20:21:32 Join damaki [0] (
20:27:14 Join linuxstb__ [0] (
20:27:22 Nick linuxstb__ is now known as linuxstb (
20:29:52 Quit linuxstb_ (Read error: 110 (Connection timed out))
20:33:43XavierGrWow the default rockbox font is too large for the H300 screen IMHO.
20:33:50XavierGrI mean too small
20:35:19preglowyes it is
20:35:29preglowit is sized for the archos screen
20:35:37preglowi like it just fine, but a lot of people probably don't
20:36:43linuxstbI think it gives a bad first impression - a nice big font as the default would be more friendly.
20:36:51Papricai dont know wtf is wrong with the bitmap load
20:36:56muesli__i suggest to delete h1xx wps from h3xx build cos they dont fit in any way
20:37:15Papricaany idea?
20:37:24lostlogicI use the 6x12 font on H3x0.
20:38:00linuxstbmuesli__: If you do that, then there will be no wps's left for the h3xx....
20:38:13 Quit muesli__ (Read error: 104 (Connection reset by peer))
20:38:26 Join muesli__ [0] (n=muesli_t@
20:38:54muesli__linuxstb sorry, my client crashed
20:39:08muesli__what was it?
20:39:15linuxstbmuesli__: If you do that, then there will be no wps's left for the h3xx....
20:39:37linuxstbBut I agree it would make sense once the WPS system is more stable, and we have colour WPSs in CVS.
20:39:44XavierGrmuesli is right. Current wps's are like sh*t in the H300 screen.
20:39:47muesli__there are loads of 24bit wps already arround
20:40:00linuxstbAre they in the patch tracker? Do they work with CVS Rockbox?
20:40:28muesli__afaik color-support is already implemented
20:41:10XavierGralso why does the build includes plugins that won't work?
20:41:39linuxstbmuesli__: Then please test against CVS, make sure the copyright on the images is OK, and ask the authors to submit them to the patch tracker.
20:42:18linuxstbXavierGr: So that people can fix them.
20:42:47muesli__copyright...i am not the author..dunno if those wps are free of copyright :o
20:43:04linuxstbmuesli__: Then ask the authors.
20:43:10XavierGrCan we have an 24bit wps in cvs rockbox now?
20:43:26linuxstbI don't see why not - if they work with CVS Rockbox.
20:43:32 Join PaulJ [0] (
20:44:39 Quit DreamTactix291 (Read error: 110 (Connection timed out))
20:44:50XavierGrhaha I just had my first Rockbox crash in H340
20:45:07XavierGrIs that still unstable? I just selected a song
20:45:28muesli__working like a charm over there
20:48:04XavierGrThe iriver jpeg viewer on iriver firmware sucks big time indeed! No zoom, it takes forever to load the pic and you can't view while playing music. (the last is a problem in rockbox too, but my patch solves that)
20:50:39 Quit matsl (Remote closed the connection)
20:51:57XavierGrmuesli does the bootlader USB mode works for you?
20:52:11 Join Bger [0] (n=Bager@
20:52:18XavierGrOh hello Bger!
20:52:42BgerXavierGr :)
20:52:58XavierGrBger does the bootlaoder USB mode works for you?
20:53:21BgerXavierGr yes, it does ...
20:53:32Bgeru're using the right port, yep ?
20:54:22XavierGrI don't think I can use the other one for that. Also it is placed on the cradle so...
20:55:16 Quit linuxstb ("Leaving")
20:55:37XavierGrFor gods sake I can't use the USB OTG function. Anone that can help me?
20:56:11muesli__XavierGr check mr
20:56:18muesli__there is a nice video
20:56:37XavierGrI think I do what the video says but I get an error!
20:56:52XavierGrOver Current! Remove device and press stop.
20:57:04muesli__not all devices are supported
20:57:19XavierGrI am trying to do that between my H300 and H100.
20:57:21 Part amiconn
20:57:57 Quit beeble (Remote closed the connection)
20:58:17 Join amiconn [0] (
21:00:58BgerXavierGr ah
21:01:09Bgeri've done this between 2 h300s...
21:01:42BgerXavierGr what's your fw version ?
21:04:10XavierGr1.29 kr
21:04:36***Saving seen data "./dancer.seen"
21:04:42XavierGrEven if I plug the USB host cable without connecting the other end I get the same error.
21:06:04XavierGrHere is what I do. I press rec and go to Browser. Then I connect the USB host cable (with nothing in the other end) and I get this message.
21:06:34Bgerthis is not good at all :(
21:06:41Bgeror the cable is shorted
21:07:31XavierGrI get the same error when I plug a standart mini b usb cable.
21:08:20XavierGrno wait, I will get the error only with the USB Host cable
21:10:12 Join freqmod [0] (
21:10:29BgerXavierGr u're sure it says " Over Current! " only with the host cabvle ?
21:11:00XavierGrI will try again.
21:11:10Bgerreset the settings
21:11:17Bgerfor a start
21:11:55XavierGrUSB OTG port USB HHOST cable: error
21:12:00amiconnXavierGr: Perhaps your USB host cable is faulty.
21:12:11amiconnI remember someone else reporting similar symptoms
21:12:25XavierGrUSB port normal cable: no error
21:12:49BgerXavierGr yep, you should try other cable...
21:13:00Bgermini-A - A
21:13:01XavierGrgreat and where can I find such cable?
21:13:14Bgeri don't know
21:13:24Bgerin fact i haven't seen such cable here ... :)
21:13:25XavierGrI haven't seen a connector like this before.
21:13:25amiconnXavierGr: If you have a multimeter, check the host cable.
21:13:48XavierGrvery difficult to check the pins in the mini adapter are too little
21:14:18BgerXavierGr see the DataSheet wiki
21:14:54Bgerit should be described here
21:15:00Bger(the mini-A pinout)
21:17:23Bgeramiconn can you have a look at the plugin's code again ? :(
21:17:27muesli__any complaints?
21:18:38Bgerdo you think that 500 bytes should be enough for the thread's stack ?
21:20:04XavierGrI think that the player is to blame not the cable.
21:21:00XavierGrAnd to think that I bought it for the USB OTG. Sh*t :(
21:21:22amiconnBger: I would just use DEFAULT_STACK_SIZE, which is 1KB
21:21:33amiconnI'll take a look in a few mins
21:21:33 Quit ssnajper ("CGI:IRC (EOF)")
21:21:40Bgeramiconn 10x
21:23:08 Join Febs [0] (
21:23:27Bgeri'm asking because there are about 100bytes of local vars in the stack
21:23:35Bgerand it stkov's with 500 bytes...
21:25:23FebsI was just reading the discussion from a short while ago between linuxstb and muesli about H300 WPS screens.
21:25:42muesli__i am just reading ur reply ;)
21:25:47amiconnBger: A stkov doesn't necessarily mean a real stack overflow
21:26:00FebsI did a WPS called iAmp300, which is based on Raymond Ho's iAmp WPS that is already in the H100 and H300 builds.
21:26:03amiconnIt can also be something else overwriting the beginning of the stack
21:26:23FebsThe iAmp300 WPS can replace the iAmp WPS in the H300 builds.
21:26:43amiconnThe thread scheduler checks the very beginning of each thread's stack during a yield(), and panics if it's no longer 0xDEADBEEF
21:26:58FebsIt's in the patch tracker if anyone with CVS commit access would take a moment to commit it to CVS.
21:26:59preglowsomeone make a retailos clone i can put on my nano, plz ;)
21:27:29FebsI have Raymond Ho's permission (in writing) to use the WPS, so there is no copyright issue.
21:28:36 Quit Sinbios (Read error: 104 (Connection reset by peer))
21:28:43 Join Sinbios [0] (
21:29:05Bgeramiconn i supposed that it's something like this
21:29:30muesli__Febs does color bmp eat up more juice at all?
21:30:06Bgeramiconn it's strange because it stkovs after 2-3 secs at most ...
21:30:13XavierGrhave to go, later all
21:30:14 Part XavierGr
21:31:15Febsmuesli, LinusN wrote recently on MR that 24bit WPS consume more battery. I'll have to take his word for it. ;)
21:31:22Bgerand that should be after the first sleep ...
21:31:30muesli__Febs ok ;)
21:31:56muesli__but how many want b/w on their color display? my guess 0.0001%
21:31:58Bgeri mean the first queue_wait_w_tmo()
21:36:40LearBger: the stack needs to be large enough for any called functions too; I guess you call a few at least...
21:37:38 Quit San (Read error: 110 (Connection timed out))
21:38:04 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
21:38:29Bgeropen, fdprintf ...close..
21:38:51amiconnopen needs > MAX_PATH bytes
21:39:29LearMeaning 1 kB sounds like a good idea...
21:40:42amiconnopen() needs somewhat above MAX_PATH, and it calls opendir, which needs something about the same amount
21:41:01amiconnThat's already >500 bytes ...
21:41:20preglowand MAX_PATH will be larger, yes?
21:41:25preglowit's not even utf8-ified yet
21:41:39 Join wire [0] (
21:41:40amiconnYes it will probably be larger in the future
21:41:46 Join beeble [0] (
21:41:46amiconnToday it's 260 bytes
21:41:53Bgerthis explains everything
21:41:53preglowit's probably going to be three times that
21:42:20Bgeri thought that i have a bug in the code i couldn't see...
21:42:34wiresomeone maybe can help me my iriver h320 with rockbox bootloader freeze in the start
21:43:24FebsTry pressing the record button immediately after turning it on. This will reset your settings to the default.
21:44:09wirethe iriver software is fine, but the rockbox stack after he write Vresion 5.....
21:46:14wirei try to upgrade the firmware but its still the same
21:52:56 Join Genre9mp3 [0] (
21:53:16 Quit Genre9mp3 (Client Quit)
21:53:32 Join ehntoo [0] (
21:55:28Febswire, did you try what I said? Upgrading the firmware will not change your settings, but my suggestion will.
21:57:19Bgerwire does the bootloader read from disk
21:57:35Bgerdoes it output things like "Model: " Checksum .....
21:57:50 Nick paugh is now known as AliasCoffee (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
21:57:51wirebut that it
21:57:57BgerResult: 0?
21:58:06wireafter Version 5
21:58:08wireone more line
21:58:13wireand thats it
21:58:20Bgerand it's ?
21:58:20 Join solexx_ [0] (
21:58:54wirei'm sorry but its not here
21:59:32wireim in trip around the world and i'm in bangkok and its not on me right now
21:59:51wirei will come tomorow with the iriver and hope u can help me then
22:00:14Bgerwire isn't it "Batt: x.xx V" ?
22:00:25markunwire: how's bangkok?
22:00:43Bgerthe next line, that is
22:00:45wirebangkok its crowdy and noisy
22:00:59wirethe islands and the north is much better
22:01:53muesli__bkk *inlove*
22:02:00muesli__dust, dirt and yummy food!!!
22:07:17wirebger yes it batt: x.xx
22:07:31wirebut then nothing happen
22:08:11preglowcan i assume my text segment is writable? :>
22:08:12Bgerthe next line is loading firmware ...
22:08:29Bgerand then it spins the disk ...
22:08:39Bgeramiconn : stkov with 1kb too ..........
22:09:16Bgerwith DEFAULT_STACK_SIZE
22:11:02 Join Domonoky [0] (
22:11:17 Join Cassandra [0] (
22:11:49*Domonoky made new powerups for brickmania :-) longer/smaller pad ..
22:12:03 Quit bluey ("Leaving")
22:12:29Domonokysource and bmps:
22:13:37Domonokythere are also rocks, for h1x0, h1x0sim and h3x0, at this location, if someone likes to test :-)
22:14:18lostlogicHey devs, do we want my numeric file sort, or should I trash it?
22:14:24 Join mymomthelush [0] (
22:14:56 Join t0mas_ [0] (
22:15:08preglowlostlogic: how is it integrated
22:15:15 Join Lear_ [0] (
22:15:32 Quit Lear_ (Client Quit)
22:16:01 Quit t0mas (Read error: 104 (Connection reset by peer))
22:16:01 Quit damaki (Read error: 101 (Network is unreachable))
22:16:01 Join damaki [0] (
22:16:06PapricaDomonoky, i will check it in a moment =]
22:16:41 Quit mymomthelush (Client Quit)
22:18:18 Quit solexx (Read error: 110 (Connection timed out))
22:18:46CassandraWhat do you mean by numeric file sort?
22:19:33preglowCassandra: sorts by track number, i guess
22:19:41preglowCassandra: for file names that start with a number
22:20:01preglowCassandra: btw, for your distributed build system idea, how would we guarantee that all systems generate the same output?
22:20:54preglowCassandra: and do you know about distcc?
22:22:00 Quit Matze41 ("Miranda IM! Smaller, Faster, Easier.")
22:22:12 Join DreamTactix291 [0] (
22:22:12wireBGER i dont think it load the firmware
22:22:32wirebut what i can do to delete all and start over
22:26:24PapricaDomonoky, the power ups works in h300 too?
22:26:39Papricai didnt see them..
22:28:23DomonokyLo and s are the new poweups..
22:28:33Papricado you have a rock for the sim?
22:29:44Domonokybuild with yesterdays CVS, hope it works :-)
22:29:58 Quit wire ()
22:30:04 Join linuxstb [0] (
22:31:10Papricamm yep now it works
22:31:15Papricai compile the simulator again
22:31:17linuxstbCassandra: Any plans to continue your installer project?
22:32:23 Quit Lear (Read error: 110 (Connection timed out))
22:32:25Domonokyrock for h1x0sim:
22:32:39 Join Lear [0] (
22:33:13LearAnyone tried using WPS with grayscale images on a H1x0?
22:34:11lostlogicpreglow: it's a new sort option in the list −− it could be made part of alpha, basically if the beginnings of both files start with numbers, they are compared by the leading number, failing over to alphabetic
22:34:44LearI seem to get random loading garbage on target... Mostly okay, but occasional problems, of a random nature.
22:35:58preglowlostlogic: i'm fine with it, get a couple of more opinions and see
22:36:32BagderI'm mildly against it
22:37:10 Join linuxstb_ [0] (
22:37:21 Quit linuxstb (Nick collision from services.)
22:37:28 Nick linuxstb_ is now known as linuxstb (
22:38:53lostlogicBagder: reasoning?
22:39:03Bagderit adds very little value
22:39:28Bagderand add a weird sort case
22:40:02lostlogic*shrug* I obviously coded it because it was useful to me because it doesn't break sorting on any of my albums, but correctly sorts several that were otherwise incorrectly sorted.
22:40:14Bagderin my book, it is in the chapter "things we can think of but don't need"
22:40:14preglowBagder: so, what do you say about distcc'ing the bleeding builds if you can find people to donate some server usage?
22:40:22preglowBagder: would save you buying a new box, eventually
22:40:34Bagdersure, that could be fun
22:40:36linuxstbHow well does distcc work over slow networks?
22:40:50lostlogicwe'll need to make the build properly parallel build compliant first
22:41:01preglowas long as the central machine has enough bw, then sure
22:41:13lostlogicand I can donate processor and bandwidth on my colocated server
22:41:22linuxstbWouldn't it be easier just to farm out individual builds in their entirety?
22:41:52preglowBagder: slasheri's already said he's going to remove it
22:41:57preglowhe just hasn't done it yet, heh
22:41:58webguest57please make it so random volunteers can help compile too :)
22:42:18Bagderok, then I'm just slow... as usual ;-)
22:42:27 Join San [0] (
22:42:31*preglow produlates Slasheri
22:42:35Cassandralinuxstb, yes. Very definite plans.
22:42:49Bagder"random" couldn't do it since we'd want some kind of knowledge the builds can be semi-trusted
22:42:49CassandraThe problem is it's kind of on hold till after I've moved house.
22:43:12Bagderif we'd use them for bleeding edge like today
22:43:18 Quit freqmod (Remote closed the connection)
22:43:25CassandraSince after getting feedback I realised that for it to be useful to the widest number of people it needed an architectural change.
22:43:34 Quit DreamTactix291 (Read error: 110 (Connection timed out))
22:44:11CassandraI keep meaning to put out an interim build that at least has iPod download support, since there's nothing really wrong with the basic functionality.
22:44:20 Join linuxstb_ [0] (
22:44:25 Quit linuxstb (Nick collision from services.)
22:44:27 Nick linuxstb_ is now known as linuxstb (
22:44:28preglowlinuxstb: well, perhaps
22:44:33preglowi dunno
22:44:36preglowdistcc exists and works
22:44:51preglowshouldn't be that hard to automate an entire build, i guess
22:44:52 Join Genre9mp3 [0] (
22:45:00lostlogicBagder: if you want a user account on my server to teach it to do some of the builds, let me know... it's a 2.8ghz p4 ht on a 100mb pipe
22:45:17CassandraI'm happy to run builds on my dev box.
22:45:36Bagderlet me ponder on it and get back
22:45:44CassandraIt's only on ADSL, but should be enough for helping with a distributed build.
22:46:10preglowwith a small amount of distcc and ccacheed boxes running, our build table should populate a great deal faster
22:46:16 Quit zhilik ("")
22:46:20 Join Thus0 [0] (
22:46:28BagderI think so too
22:46:50lostlogicwould take away the excuses we use for committing and running :-P
22:47:15preglowand we dont't switch compiler tools so often that it would be a pain
22:47:46preglowand if it was ever to be a problem, distcc can be switched off until it's fixed
22:47:50lostlogicBagder: what do you think of incorporating the numeric sort into the normal alpha sort only if both filenames begin with numeric, ro is that even worse?
22:48:05safetydannearly done with a simple eq UI, now I just need to write a parser for eqs files
22:48:22 Quit Domonoky (Read error: 104 (Connection reset by peer))
22:48:29safetydanBasically you'll have "Browse Equalizer Settings" in the sound settings menu
22:48:55safetydanyou'll have to hand write the files for the moment
22:49:55LinusNsafetydan: you should be able to reuse the same parser as we use for the .cfg files
22:52:26preglowsafetydan: have you figured out how to call the eq_*_coefs functions properly?
22:52:49LearHah, another grayscale bmp loader bug found. :)
22:54:12preglowBagder: still no idea about the eq_sf.S bug?
22:54:30Bagderno, and I haven't had much time to investigate
22:54:32LearSame as the old one really, but on another place...
22:54:33LinusNLear: it's that Linus fellow, he keeps screwing things up!
22:55:02LinusNpreglow: what bug?
22:55:09LearWell, in this case I understand if mistakes are made. Lots of different cases to test... :)
22:55:18preglowLinusN: .S files in apps/ aren't compiled properly, they're just passed straight to the linker process
22:55:26preglowLinusN: and they're deleted on make clean...
22:55:35Bagder.c is replaced by .o
22:55:37CassandraOh, that reminds me - is 4 colour greyscale for the H1xx working now?
22:55:41Bagdernot taking case of .S
22:55:46preglowi did that and had my work deleted for me the other day
22:55:49Bagderbut that's not that only error
22:55:52LearSort of; about to fix a bug in the bmp loader...
22:57:08LearThere's one image that doesn't seem to get light gray as it should though...
22:58:28 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
22:58:41preglowBagder: how is it with arm, btw? is it beneficial to interleave instructions with loads/stores, or just put loads/stores in great bunches
22:59:05safetydanpreglow: I haven't got a clue about Q and cutoffs and so forth, but the functions look straightfoward so I'll just through numbers at them.
22:59:07BagderI'm not sure
22:59:13safetydananyway gotta get to sleep
22:59:14 Part safetydan ("Leaving")
22:59:39preglowsafetydan: wait...
23:00:15 Quit Bger ("[BX] It's a huge shit sandwich and we're all gonna have to take a bite.")
23:01:58LearWait a minute, I'm using the wrong colors! Silly me... :)
23:02:24Bagderlostlogic / Cassandra: you have all the cross-compilers present? can you run distccd and tell me your IP addresses (possibly privately) and I'd do some testing
23:02:43Bagdersetup to only allow
23:03:09lostlogicBagder: hmm... I'll have to change my setup −− they are not installed for systemwide currently
23:03:22Bagderthere's no hurry
23:03:34BagderI'll run some tests with my own homebox as well
23:04:37CassandraLear: Good to know. I'll convert boxes to proper greyscale when I have a bit of time.
23:04:39***Saving seen data "./dancer.seen"
23:05:09preglowhooray for me! hooray for preglow!
23:05:21preglowanother award-winning idea
23:05:26preglowyou guys should give me an award
23:05:44*Bagder hands preglow an award he had lying around on the floor
23:05:57*Cassandra awards preglow more Rockbox work.
23:06:05LinusNpreglow: i have fixed the .S makefile issue
23:06:08*preglow grabs defensively
23:06:12 Quit damaki (Read error: 104 (Connection reset by peer))
23:06:17preglowLinusN: hooray!
23:06:23Bagdernow LinusN can have an award too!
23:06:26*Cassandra wants to do a 5x3 Rockbox font for shits and giggles at some point.
23:06:39preglowand i'm going to convert the standard dos font
23:06:49CassandraMostly to see what it'd look like on the Nano.
23:07:02preglowit occured to me that i now have a colour screen unit, which alone is reason enough to code an ansi viewer
23:07:20LearCassandra: Note that dark gray is very close to black in the Win32 sim at least... :)
23:07:25 Join damaki [0] (
23:07:45CassandraLear. Sounds like a bug in the sim.
23:08:03Bagderor a bad monitor ;-)
23:08:08CassandraThe other thing I'd really like to do is port one of the Z-machine interpreters.
23:08:17CassandraTo a rockbox plugin, that is.
23:08:22LearIt isn't black; if I zoom it up a bit it is clearly different from black.
23:08:46LearThis monitor got good marks from Toms Hardware at least, so it shouldn't be that bad... :/
23:09:05Bagderthrow it away and buy a new!
23:09:29preglowi'm doing eq_filter() for arm for shits and giggles
23:10:13 Quit ghode|afk (Read error: 104 (Connection reset by peer))
23:10:29 Quit damaki (Remote closed the connection)
23:10:56preglowi don't like this load/store architecture thing at all
23:11:02preglowbut apart from that, arm asm is really nice
23:11:58 Quit gromit` (Remote closed the connection)
23:13:00CassandraEq confuses me. It's kind of like going to an art gallery and throwing paint on the canvases,
23:13:34Paul_The_NerdCassandra: Sometimes though, you're just trying to compensate for bad speakers/earphones.
23:13:44preglowbut i never use it myself, though
23:13:58Paul_The_NerdI think profiles for various common earphones would be kinda neat to include.
23:14:02 Join gromit` [0] (n=gromit`
23:14:11preglowPaul_The_Nerd: as opposed to Rock/Jazz etc, it would actually be useful
23:14:15preglowPaul_The_Nerd: and i agree completely
23:15:37Learpaul_the_nerd: getting good bass from earbuds isn't that easy, it seems.
23:16:11preglowit suddenly occured to me that the eq has no way of lowering general volume
23:16:25preglowand that will most certainly be needed
23:16:34lostlogicreplaygain preamp?
23:16:39beebleLear: good inears for the street and akg or sennheiser for homeuse
23:16:48preglowlostlogic: well, yeah, but that only works on replaygain files
23:16:48CassandraPaul: Or even good ones. My ER-4Ps are lovely, but the bass needs driving really hard.
23:16:52preglowlostlogic: but something just like that, yes
23:17:23beebleer-4p arent designed for bass ;)
23:17:34lostlogicBagder: are the iPod bleeding edge builds done on gcc4?
23:18:13lostlogicBagder: need arm-gcc4.0.2 and m68k/sh1-gcc-3.4.5 and all 3 binutils-2.16.1?
23:18:14Paul_The_NerdIs there any objective data on such things, or at least a way to objectively measure where the flaws are, or would it just be a matter of subjectively trying to compensate?
23:18:19LearHm... Getting a decent WPS here I think. Lower half needs more work though...
23:18:23Cassandrabeeble: Nope, although I do wonder if the Shure E5c's do a better job.
23:19:09CassandraTheoretically you ought to be able to model EQ settings on the frequency response curve of the headphones. I think.
23:19:09Bagderlostlogic: binutils 2.16.* is needed for the m68k case, the others aren't that strict iirc
23:19:50beebleCassandra: do you realy need full isolation? if not, sony ex71sl is a very good cheat alternative
23:20:21beebleand there are even avail. in ipod comapt white ;)
23:21:16beeblethey are very cheap for the good sound, just the cardle is a little bit fragil
23:22:13 Join ep0ch [0] (n=ep0ch@
23:23:47ep0chspeaking of preamp, i have an idea how i'd like to see it work in future...
23:23:59preglowspill it
23:24:01ep0chbasically strip out the preamp from replay gain
23:24:14preglowyes, agreed
23:24:17ep0chhave one place for replay gain
23:24:40Learbeeble: I think it's a pair of EX71:s I have. Not very good compared to Shure or Etymotics...
23:24:58LearTrue, there's a price difference...
23:25:12ep0chthe new preamp should has several settings in it, each one adds to the total preamp value
23:25:24FebsAs quickly as EX-71s wear out, you're better off getting a pair of Shure E2s. In the long term, they're probably less expensive.
23:25:27preglowseveral? why?
23:25:38ep0chso if replay gain enabled add +3db, if crossfeed enabled +4db, if no replaygain +4db etc
23:25:48beebleLear: in my tests they where quite the same, just the shure or ety have full isolations which can be _very_ usefull in public places
23:26:00LearFebs: not for me. Cables start to break after about 7 months. No good deal. :(
23:26:02beeblebut it can also be your death ;)
23:26:04ep0chget the idea?
23:26:27FebsCables on Shures? Or Sony? IIRC, Shure has a 2 year warranty.
23:26:27Cassandrabeeble, trust me - when you've heard ER-4ps you don't want to go back to cheap headphones.
23:26:40ep0chthe issue at the moment is that enabling crossfeed reduces volume, this could overcome that
23:26:46beeblei tested them
23:26:48CassandraPlus they are lovely and quiet on public transport. Really reduces the stress of travel.
23:27:16LearFebs: On Shures (E2c). I got one warranty replacement. Same story. So I gave up.
23:27:17beebleas i say, if you depend on isolation -> ety (could get my hand on shure)
23:27:32CassandraI've had my ER-4ps for 2 years. Cable is still fine. You get what you pay for.
23:27:59FebsLear: bummer. Fortunately, I haven't had any such problem with any of my Shure IEMs.
23:28:01LearThey cost like three times more, or what is it?
23:28:30LearAnd if they don't have the cables running over the ears, that should reduce the problem a lot.
23:28:50beebleand i must say, i will stay with my akg 501 at home
23:28:52FebsIn the U.S., EX-71s are about $30-35. Shure E2 about $75, so yeah, 2 to 2.5 times.
23:29:38LearFebs: I was replying to Cassandra... :)
23:30:25FebsOops! :)
23:31:38preglowFebs: started on your ui mapping endeavour, btw?
23:31:55slimxhey all :)
23:31:58FebsYeah, but not much progress so far.
23:32:58slimxcan someone explain me what's related to the -32 ata_init () return code ?
23:33:04FebsI spent a plane ride mapping out all of the menus.
23:33:57slimxi got this error on the ipod3g port
23:34:19 Join TCK- [0] (
23:34:30 Join XavierGr [0] (
23:35:12 Join ismo [0] (
23:35:35 Join matsl [0] (
23:36:05 Quit XavierGr (Read error: 104 (Connection reset by peer))
23:36:49 Join XavierGr [0] (
23:37:20 Quit ender` (" printk("autofs: Out of inode numbers -- what the heck did you do??\n"); -- /usr/src/linux/fs/autofs/root.c")
23:39:26linuxstbslimx: The check_registers() function is failing.
23:39:59 Quit Lear ("Chatzilla 0.9.70 [Firefox 1.5/2005111116]")
23:40:00 Join ender` [0] (i=ychat@
23:40:28 Quit ismo_ (Read error: 113 (No route to host))
23:40:35linuxstbslimx: Have you changed the register addresses for the PP5002?
23:40:45 Quit ep0ch ("Kopete 0.11 :")
23:41:59 Quit NicoFR ("Konversation terminated!")
23:44:18 Quit Thus0 (Connection timed out)
23:49:30Midgey34has any iPod user tried blackjack
23:49:37Slasheripreglow: i will remove it tomorrow morning unless somebody else comes first and try to determine ways to support the early dircache thing.. :)
23:50:10preglowSlasheri: aight
23:50:16Paul_The_NerdMidgey34: I have, on Nano
23:50:20preglowi've never reverted anything in my life, so i'll keep it out of it
23:50:35preglowwould probably end up deleting whole dirs
23:50:52slimxi guess yes linuxstb
23:51:06Midgey34Paul_The_Nerd: how did the key definitions feel
23:51:06 Quit Sinbios (Read error: 104 (Connection reset by peer))
23:51:10 Join Sinbios [0] (
23:51:12Midgey34I would guess they could use work
23:51:13slimxbut tell me where i must check
23:51:31XavierGrdoes anyone know how to pen an H300 here?
23:51:37Paul_The_NerdMidgey34: Honestly, I think the center button should be "hit"
23:51:43XavierGrI removed all the screws but I can't open it up.
23:52:01muesli__its a pain in the ass
23:52:06linuxstbslimx: If you want, I can commit an ata.c to CVS that should work for you. I know where to look in ipodlinux for the info.
23:52:13muesli__use your finger nails
23:52:59XavierGrhmm I will try again.
23:53:50Genre9mp3wow XavierGr.....i'm very impressed....Try to open a brand new H340 from day 1.... ;-)
23:54:03slimxit would be great
23:56:44Rob2222XavierGr: I can help
23:56:53Rob2222XavierGr: Do you opened it, yet?
23:57:03linuxstbslimx: Try this: - as far as I can see, that should work.
23:57:11XavierGrNo I have removed teh screws
23:57:23XavierGr1 in each side except the bottom which has 2, right?
23:57:32Rob2222XavierGr: right.
23:57:48slimxthx linuxstb
23:57:48Rob2222XavierGr: Now look at the bottom.
23:57:48BHSPitLappysoak it in hot water
23:58:06linuxstbslimx: Wait a minute - I'll upload a better version...
23:58:26Rob2222XavierGr: Try to get a fingernail between rear cover and middle pure black plastic.
23:58:32preglowPrefetch abort at 0xDEADBEEE
23:58:34Rob2222XavierGr: At the bottom
23:58:38*slimx is waiting
23:58:39preglownice hint at what i did wrong
23:58:58BHSPitLappypreglow: ever get the ipod out?

Previous day | Next day